{ "cells": [ { "attachments": {}, "cell_type": "markdown", "metadata": { "id": "55AbwMnJGvs7" }, "source": [ "# EGB_res网络可视化\n", "\n", "**目标:**\n", "\n", "\n", "1. 学习神经网络在训练过程中,隐含层的特征变化机制\n", "2. 学习端到端梯度提升算法(End to end gradient boosting, EGB)\n", "\n", "代码中的基学习器为残差网络(residual network, res)\n", "\n", "\n", "**参考:**\n", "\n", "葛家驿,杨乃森,唐宏,徐朋磊,纪超.端到端的梯度提升网络分类过程可视化[J].信号处理,2022,38(02):355-366.DOI:10.16798/j.issn.1003-0530.2022.02.015.\n", "\n" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## 研究准备" ] }, { "cell_type": "markdown", "metadata": { "id": "90bw-7WJL55q" }, "source": [ "### **环境配置**" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "TUmq2vFS_DNN", "outputId": "0310fe8b-417a-4f20-a139-63b6aa41d63b" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Populating the interactive namespace from numpy and matplotlib\n" ] } ], "source": [ "'''\n", "端到端梯度提升模型分类过程可视化\n", "每个基分类器是残差网络\n", "'''\n", "\n", "import numpy as np\n", "from sklearn.preprocessing import StandardScaler\n", "import tensorflow as tf\n", "from tensorflow import keras\n", "from tensorflow.keras import backend as K\n", "from tensorflow.keras import layers\n", "from tensorflow.keras.callbacks import CSVLogger\n", "from sklearn.preprocessing import MinMaxScaler\n", "from matplotlib.colors import ListedColormap\n", "from matplotlib import cm\n", "import matplotlib as mpl\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "from sklearn.datasets import make_circles #同心圆数据\n", "from sklearn.model_selection import train_test_split\n", "import sys\n", "sys.setrecursionlimit(500000)\n", "import imageio\n", "%pylab inline " ] }, { "cell_type": "markdown", "metadata": { "id": "9XXBVJnVL71I" }, "source": [ "### **设置模拟数据**" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "6VTxCvvH_K03" }, "outputs": [], "source": [ "n_samples = 1000 #样本点数\n", "X, y = make_circles(n_samples=1000,factor=.4,noise=.06,random_state=0) #生成同心圆数据\n", "test_size = 0.5\n", "\n", "#划分训练集、测试集\n", "X_train, X_test, Y_train, Y_test = train_test_split(X, y, test_size=test_size, random_state=2)\n", "\n", "c,r = np.mgrid[[slice(X.min()- .2,X.max() + .2,50j)]*2]\n", "p = np.c_[c.flat,r.flat]\n", "\n", "ss = StandardScaler().fit(X_train)\n", "X = ss.transform(X)\n", "p = ss.transform(p)\n", "X_train = ss.transform(X_train)\n", "X_test = ss.transform(X_test)" ] }, { "cell_type": "markdown", "metadata": { "id": "8lJEaUD2L8-1" }, "source": [ "### **配置绘图环境**" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 600 }, "id": "fu3TqOmc_SGR", "outputId": "8616854b-5aa5-4e8e-d673-cf3047a8c7e8" }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "#设置画布大小和颜色\n", "fig = plt.figure(figsize = (9,3))\n", "top = cm.get_cmap('Oranges_r', 512)\n", "bottom = cm.get_cmap('Blues', 512)\n", "newcolors = np.vstack((top(np.linspace(0.55, 1, 512)),\n", " bottom(np.linspace(0, 0.75, 512))))\n", "cm_bright = ListedColormap(newcolors, name='OrangeBlue')\n", "\n", "#训练数据可视化\n", "plt.subplot(121)\n", "m1 = plt.scatter(*X_train.T,c = Y_train,cmap = cm_bright,edgecolors='white',s = 20,linewidths = 0.5)\n", "plt.title(f'train data ({int(n_samples*(1-test_size))} points)')\n", "plt.axis('equal')\n", "\n", "#测试数据可视化\n", "plt.subplot(122)\n", "m2 = plt.scatter(*X_test.T,c = Y_test,cmap = cm_bright,edgecolors='white',s = 20,linewidths = 0.5);\n", "plt.title(f'test data ({int(n_samples*test_size)} points)')\n", "plt.axis('equal')\n", "ax = fig.get_axes()\n", "plt.colorbar(ax = ax)\n", "#plt.savefig(f'data_{n_samples}_points.png')\n", "#plt.savefig(f'data_{n_samples}_points.pdf')\n", "plt.show()\n", "\n", "#全部数据可视化\n", "fig = plt.figure(figsize = (7,6))\n", "plt.scatter(*X.T,c = y,cmap = cm_bright,edgecolors='white',s = 20,linewidths = 0.5)\n", "plt.title(f'Raw data ({n_samples} points)')\n", "plt.axis('equal')\n", "#plt.savefig(f'Raw data ({n_samples} points)')\n", "#plt.savefig(f'Raw data ({n_samples} points).pdf')\n", "plt.axis('equal')\n", "#plt.colorbar(ax = ax)\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": { "id": "Je8SuWBkL-Tn" }, "source": [ "### **配置数据与损失函数**" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "UFHq70m-_T2m" }, "outputs": [], "source": [ "num_classes=2 #设置类别数\n", "y_train=keras.utils.to_categorical(Y_train,num_classes) #类别标签转换为onehot编码\n", "y_test=keras.utils.to_categorical(Y_test,num_classes)\n", "#定义损失曲线绘制函数\n", "\n", "def plot_loss_accuracy(history, title_text, file_name):\n", "\n", " fig, ax1 = plt.subplots()\n", " ax2 = ax1.twinx()\n", " ax1.set_xlabel(\"Epoch\")\n", " ax1.set_ylabel(\"Accuracy\")\n", " ax2.set_ylabel(\"Loss\")\n", " \n", " #ax1.set_ylim(-0.01,1.01)\n", " ax1.plot(history.epoch,\n", " history.history['accuracy'],\n", " label=\"Training Accuracy\")\n", " ax1.plot(history.epoch,\n", " history.history['val_accuracy'],\n", " linestyle='--',\n", " label=\"Test Accuracy\")\n", " \n", " #ax2.set_ylim(-0.01,1.01)\n", " ax2._get_lines.prop_cycler = ax1._get_lines.prop_cycler \n", " \n", " ax2.plot(history.epoch, \n", " history.history['loss'], \n", " label=\"Training Loss\")\n", " ax2.plot(history.epoch,\n", " history.history['val_loss'],\n", " linestyle='--',\n", " label=\"Test Loss\")\n", "\n", " ax1.legend()\n", " ax2.legend()\n", " plt.suptitle(title_text)\n", " plt.savefig(file_name)" ] }, { "cell_type": "markdown", "metadata": { "id": "PNzM9oXyMBem" }, "source": [ "### **配置基于残差网络的端到端梯度提升模型**" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "l3fHLYSX_fET" }, "outputs": [], "source": [ "#定义残差块\n", "\n", "def ResMLP_Block(name_of_classifiers,\n", " num_of_res_blocks,\n", " num_hidden_layers_of_res_block,\n", " num_neurons_of_hidden_layer,\n", " inputs):\n", " \n", " x = inputs\n", " for i in range(num_of_res_blocks):\n", " inputs = x\n", " for j in range(num_hidden_layers_of_res_block):\n", " x = layers.Dense(num_neurons_of_hidden_layer,\n", " activation=tf.nn.relu,\n", " name = f'{name_of_classifiers}th-clf_{i}th-block_{j}th-hidden')(x)\n", " x = layers.Dense(2, name = f'{name_of_classifiers}th-clf_{i}th-resBlock_linear')(x)\n", " x = layers.Add(name = f'{name_of_classifiers}th-clf_{i}th-resBlock_Add')([x,inputs])\n", " \n", " return x\n", "#构建残差基学习器\n", "\n", "def build_res_model(name_of_classifiers,\n", " num_of_res_blocks,\n", " num_hidden_layers_of_res_block,\n", " num_neurons_of_hidden_layer,\n", " inputs,\n", " num_of_classes=2):\n", " x = ResMLP_Block(name_of_classifiers,\n", " num_of_res_blocks,\n", " num_hidden_layers_of_res_block,\n", " num_neurons_of_hidden_layer,\n", " inputs)\n", " outputs = x\n", " #outputs = layers.Dense(num_of_classes,name=f'{name_of_classifiers}th-clf_logits')(x)\n", " res_model = keras.Model(inputs, outputs)\n", " return res_model\n", "#构建EGB网络\n", "\n", "def build_boosting_model(classifiers,\n", " name_of_classifiers,\n", " num_of_res_blocks,\n", " num_hidden_layers_of_res_block,\n", " num_neurons_of_hidden_layer,\n", " inputs):\n", " model_logits = build_res_model(name_of_classifiers,\n", " num_of_res_blocks,\n", " num_hidden_layers_of_res_block,\n", " num_neurons_of_hidden_layer,\n", " inputs)\n", " classifiers.append(model_logits)\n", " if len(classifiers)>1:\n", " res_boost_model = layers.Add(name='classifiers_Add')([item.outputs[0] for item in classifiers])\n", " else:\n", " res_boost_model = model_logits.outputs[0]\n", " outputs = layers.Dense(2, activation='softmax',name = 'activation')(res_boost_model)\n", " #print(outputs)\n", " boosting_model = keras.Model(inputs, outputs)\n", " return boosting_model,model_logits" ] }, { "cell_type": "markdown", "metadata": { "id": "xM8q7rqNO3_G" }, "source": [ "### **定义模型训练方式**" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "qpS5BJZb_jXe" }, "outputs": [], "source": [ "#定义模型训练\n", "\n", "def train_ResBoost_model(number_of_weak_classifiers,\n", " num_of_res_blocks,\n", " num_hidden_layers_of_res_block,\n", " num_neurons_of_hidden_layer,\n", " batch_size,\n", " epochs):\n", " \n", " classifiers = []\n", " history = []\n", " boosting_models = []\n", " \n", " inputs = keras.Input(shape=(2, ))\n", " \n", " #每次叠加一个基分类器拟合训练数据\n", " for n_th_weak in range(number_of_weak_classifiers):\n", " name_of_classifiers = n_th_weak\n", " boosting_model,model_logits=build_boosting_model(classifiers,\n", " name_of_classifiers,\n", " num_of_res_blocks,\n", " num_hidden_layers_of_res_block,\n", " num_neurons_of_hidden_layer,\n", " inputs)\n", "\n", " boosting_model.compile(loss=keras.losses.categorical_crossentropy,\n", " optimizer=keras.optimizers.Adam(lr=3e-4),\n", " metrics=['accuracy'])\n", " #csv_logger = CSVLogger(f'training_{name_of_classifiers}.log') #保存训练日志\n", " new_history = boosting_model.fit(X_train,\n", " y_train,\n", " batch_size=batch_size,\n", " epochs=epochs,\n", " verbose=2,\n", " #callbacks=[csv_logger],\n", " validation_data=(X_test, y_test))\n", " \n", " #在训练新的基分类器时,所有前面的基分类器每层的参数冻结\n", " for layer in classifiers[-1].layers:\n", " layer.trainable = False\n", " history.append(new_history)\n", " boosting_models.append(boosting_model)\n", " #model_logits.save_weights(f\"single_classifier_{name_of_classifiers}.h5\") #保存基分类器\n", " print('第{}个弱分类器训练完毕'.format(n_th_weak+1))\n", " \n", " #boosting_model.save(f'boosting_model_with_{number_of_weak_classifiers}_classifiers.h5') #保存模型\n", " return boosting_model,boosting_models,history" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## 进行实验" ] }, { "cell_type": "markdown", "metadata": { "id": "mdFYF5MpO5ey" }, "source": [ "### **设置训练参数**" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "prhK5dJt_lIr", "outputId": "89ac9bfe-0cb9-406d-b079-d206e599e721" }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/usr/local/lib/python3.7/dist-packages/keras/optimizers/optimizer_v2/adam.py:110: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n", " super(Adam, self).__init__(name, **kwargs)\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/400\n", "16/16 - 1s - loss: 1.2555 - accuracy: 0.5040 - val_loss: 1.3123 - val_accuracy: 0.4860 - 1s/epoch - 67ms/step\n", "Epoch 2/400\n", "16/16 - 0s - loss: 1.2451 - accuracy: 0.5040 - val_loss: 1.3009 - val_accuracy: 0.4860 - 65ms/epoch - 4ms/step\n", "Epoch 3/400\n", "16/16 - 0s - loss: 1.2351 - accuracy: 0.5000 - val_loss: 1.2891 - val_accuracy: 0.4860 - 110ms/epoch - 7ms/step\n", "Epoch 4/400\n", "16/16 - 0s - loss: 1.2250 - accuracy: 0.4960 - val_loss: 1.2779 - val_accuracy: 0.4900 - 187ms/epoch - 12ms/step\n", "Epoch 5/400\n", "16/16 - 0s - loss: 1.2153 - accuracy: 0.4940 - val_loss: 1.2669 - val_accuracy: 0.4960 - 172ms/epoch - 11ms/step\n", "Epoch 6/400\n", "16/16 - 0s - loss: 1.2059 - accuracy: 0.5000 - val_loss: 1.2556 - val_accuracy: 0.4920 - 118ms/epoch - 7ms/step\n", "Epoch 7/400\n", "16/16 - 0s - loss: 1.1969 - accuracy: 0.4960 - val_loss: 1.2441 - val_accuracy: 0.4900 - 146ms/epoch - 9ms/step\n", "Epoch 8/400\n", "16/16 - 0s - loss: 1.1870 - accuracy: 0.4940 - val_loss: 1.2343 - val_accuracy: 0.4940 - 205ms/epoch - 13ms/step\n", "Epoch 9/400\n", "16/16 - 0s - loss: 1.1782 - accuracy: 0.4960 - val_loss: 1.2238 - val_accuracy: 0.4920 - 141ms/epoch - 9ms/step\n", "Epoch 10/400\n", "16/16 - 0s - loss: 1.1697 - accuracy: 0.4940 - val_loss: 1.2132 - val_accuracy: 0.4880 - 170ms/epoch - 11ms/step\n", "Epoch 11/400\n", "16/16 - 0s - loss: 1.1606 - accuracy: 0.4920 - val_loss: 1.2038 - val_accuracy: 0.4860 - 131ms/epoch - 8ms/step\n", "Epoch 12/400\n", "16/16 - 0s - loss: 1.1524 - accuracy: 0.4880 - val_loss: 1.1940 - val_accuracy: 0.4820 - 150ms/epoch - 9ms/step\n", "Epoch 13/400\n", "16/16 - 0s - loss: 1.1441 - accuracy: 0.4840 - val_loss: 1.1842 - val_accuracy: 0.4860 - 157ms/epoch - 10ms/step\n", "Epoch 14/400\n", "16/16 - 0s - loss: 1.1360 - accuracy: 0.4840 - val_loss: 1.1746 - val_accuracy: 0.4820 - 125ms/epoch - 8ms/step\n", "Epoch 15/400\n", "16/16 - 0s - loss: 1.1281 - accuracy: 0.4860 - val_loss: 1.1654 - val_accuracy: 0.4820 - 147ms/epoch - 9ms/step\n", "Epoch 16/400\n", "16/16 - 0s - loss: 1.1202 - accuracy: 0.4860 - val_loss: 1.1566 - val_accuracy: 0.4800 - 149ms/epoch - 9ms/step\n", "Epoch 17/400\n", "16/16 - 0s - loss: 1.1126 - accuracy: 0.4860 - val_loss: 1.1483 - val_accuracy: 0.4780 - 152ms/epoch - 10ms/step\n", "Epoch 18/400\n", "16/16 - 0s - loss: 1.1053 - accuracy: 0.4860 - val_loss: 1.1394 - val_accuracy: 0.4800 - 222ms/epoch - 14ms/step\n", "Epoch 19/400\n", "16/16 - 0s - loss: 1.0985 - accuracy: 0.4860 - val_loss: 1.1303 - val_accuracy: 0.4840 - 116ms/epoch - 7ms/step\n", "Epoch 20/400\n", "16/16 - 0s - loss: 1.0906 - accuracy: 0.4840 - val_loss: 1.1227 - val_accuracy: 0.4840 - 140ms/epoch - 9ms/step\n", "Epoch 21/400\n", "16/16 - 0s - loss: 1.0836 - accuracy: 0.4860 - val_loss: 1.1150 - val_accuracy: 0.4860 - 206ms/epoch - 13ms/step\n", "Epoch 22/400\n", "16/16 - 0s - loss: 1.0770 - accuracy: 0.4840 - val_loss: 1.1066 - val_accuracy: 0.4860 - 152ms/epoch - 9ms/step\n", "Epoch 23/400\n", "16/16 - 0s - loss: 1.0703 - accuracy: 0.4800 - val_loss: 1.0987 - val_accuracy: 0.4840 - 135ms/epoch - 8ms/step\n", "Epoch 24/400\n", "16/16 - 0s - loss: 1.0635 - accuracy: 0.4800 - val_loss: 1.0915 - val_accuracy: 0.4840 - 142ms/epoch - 9ms/step\n", "Epoch 25/400\n", "16/16 - 0s - loss: 1.0572 - accuracy: 0.4800 - val_loss: 1.0840 - val_accuracy: 0.4840 - 170ms/epoch - 11ms/step\n", "Epoch 26/400\n", "16/16 - 0s - loss: 1.0506 - accuracy: 0.4800 - val_loss: 1.0768 - val_accuracy: 0.4840 - 140ms/epoch - 9ms/step\n", "Epoch 27/400\n", "16/16 - 0s - loss: 1.0445 - accuracy: 0.4800 - val_loss: 1.0694 - val_accuracy: 0.4820 - 127ms/epoch - 8ms/step\n", "Epoch 28/400\n", "16/16 - 0s - loss: 1.0383 - accuracy: 0.4800 - val_loss: 1.0626 - val_accuracy: 0.4820 - 166ms/epoch - 10ms/step\n", "Epoch 29/400\n", "16/16 - 0s - loss: 1.0323 - accuracy: 0.4760 - val_loss: 1.0559 - val_accuracy: 0.4820 - 169ms/epoch - 11ms/step\n", "Epoch 30/400\n", "16/16 - 0s - loss: 1.0265 - accuracy: 0.4760 - val_loss: 1.0490 - val_accuracy: 0.4880 - 177ms/epoch - 11ms/step\n", "Epoch 31/400\n", "16/16 - 0s - loss: 1.0207 - accuracy: 0.4760 - val_loss: 1.0422 - val_accuracy: 0.4860 - 147ms/epoch - 9ms/step\n", "Epoch 32/400\n", "16/16 - 0s - loss: 1.0148 - accuracy: 0.4740 - val_loss: 1.0362 - val_accuracy: 0.4820 - 166ms/epoch - 10ms/step\n", "Epoch 33/400\n", "16/16 - 0s - loss: 1.0092 - accuracy: 0.4720 - val_loss: 1.0299 - val_accuracy: 0.4800 - 164ms/epoch - 10ms/step\n", "Epoch 34/400\n", "16/16 - 0s - loss: 1.0038 - accuracy: 0.4720 - val_loss: 1.0235 - val_accuracy: 0.4780 - 154ms/epoch - 10ms/step\n", "Epoch 35/400\n", "16/16 - 0s - loss: 0.9982 - accuracy: 0.4720 - val_loss: 1.0175 - val_accuracy: 0.4760 - 132ms/epoch - 8ms/step\n", "Epoch 36/400\n", "16/16 - 0s - loss: 0.9931 - accuracy: 0.4700 - val_loss: 1.0107 - val_accuracy: 0.4740 - 149ms/epoch - 9ms/step\n", "Epoch 37/400\n", "16/16 - 0s - loss: 0.9875 - accuracy: 0.4720 - val_loss: 1.0046 - val_accuracy: 0.4760 - 174ms/epoch - 11ms/step\n", "Epoch 38/400\n", "16/16 - 0s - loss: 0.9823 - accuracy: 0.4720 - val_loss: 0.9988 - val_accuracy: 0.4780 - 170ms/epoch - 11ms/step\n", "Epoch 39/400\n", "16/16 - 0s - loss: 0.9772 - accuracy: 0.4760 - val_loss: 0.9934 - val_accuracy: 0.4780 - 111ms/epoch - 7ms/step\n", "Epoch 40/400\n", "16/16 - 0s - loss: 0.9723 - accuracy: 0.4760 - val_loss: 0.9876 - val_accuracy: 0.4780 - 185ms/epoch - 12ms/step\n", "Epoch 41/400\n", "16/16 - 0s - loss: 0.9672 - accuracy: 0.4760 - val_loss: 0.9823 - val_accuracy: 0.4800 - 156ms/epoch - 10ms/step\n", "Epoch 42/400\n", "16/16 - 0s - loss: 0.9628 - accuracy: 0.4740 - val_loss: 0.9762 - val_accuracy: 0.4800 - 147ms/epoch - 9ms/step\n", "Epoch 43/400\n", "16/16 - 0s - loss: 0.9575 - accuracy: 0.4740 - val_loss: 0.9712 - val_accuracy: 0.4800 - 227ms/epoch - 14ms/step\n", "Epoch 44/400\n", "16/16 - 0s - loss: 0.9531 - accuracy: 0.4720 - val_loss: 0.9654 - val_accuracy: 0.4820 - 151ms/epoch - 9ms/step\n", "Epoch 45/400\n", "16/16 - 0s - loss: 0.9483 - accuracy: 0.4720 - val_loss: 0.9606 - val_accuracy: 0.4800 - 103ms/epoch - 6ms/step\n", "Epoch 46/400\n", "16/16 - 0s - loss: 0.9438 - accuracy: 0.4720 - val_loss: 0.9554 - val_accuracy: 0.4820 - 133ms/epoch - 8ms/step\n", "Epoch 47/400\n", "16/16 - 0s - loss: 0.9393 - accuracy: 0.4720 - val_loss: 0.9502 - val_accuracy: 0.4820 - 130ms/epoch - 8ms/step\n", "Epoch 48/400\n", "16/16 - 0s - loss: 0.9347 - accuracy: 0.4740 - val_loss: 0.9458 - val_accuracy: 0.4860 - 151ms/epoch - 9ms/step\n", "Epoch 49/400\n", "16/16 - 0s - loss: 0.9306 - accuracy: 0.4680 - val_loss: 0.9405 - val_accuracy: 0.4840 - 148ms/epoch - 9ms/step\n", "Epoch 50/400\n", "16/16 - 0s - loss: 0.9260 - accuracy: 0.4700 - val_loss: 0.9356 - val_accuracy: 0.4820 - 150ms/epoch - 9ms/step\n", "Epoch 51/400\n", "16/16 - 0s - loss: 0.9218 - accuracy: 0.4700 - val_loss: 0.9306 - val_accuracy: 0.4820 - 140ms/epoch - 9ms/step\n", "Epoch 52/400\n", "16/16 - 0s - loss: 0.9176 - accuracy: 0.4720 - val_loss: 0.9257 - val_accuracy: 0.4820 - 164ms/epoch - 10ms/step\n", "Epoch 53/400\n", "16/16 - 0s - loss: 0.9134 - accuracy: 0.4720 - val_loss: 0.9214 - val_accuracy: 0.4820 - 134ms/epoch - 8ms/step\n", "Epoch 54/400\n", "16/16 - 0s - loss: 0.9094 - accuracy: 0.4720 - val_loss: 0.9168 - val_accuracy: 0.4820 - 258ms/epoch - 16ms/step\n", "Epoch 55/400\n", "16/16 - 0s - loss: 0.9053 - accuracy: 0.4700 - val_loss: 0.9123 - val_accuracy: 0.4820 - 130ms/epoch - 8ms/step\n", "Epoch 56/400\n", "16/16 - 0s - loss: 0.9014 - accuracy: 0.4720 - val_loss: 0.9077 - val_accuracy: 0.4840 - 160ms/epoch - 10ms/step\n", "Epoch 57/400\n", "16/16 - 0s - loss: 0.8975 - accuracy: 0.4720 - val_loss: 0.9033 - val_accuracy: 0.4840 - 214ms/epoch - 13ms/step\n", "Epoch 58/400\n", "16/16 - 0s - loss: 0.8934 - accuracy: 0.4720 - val_loss: 0.8994 - val_accuracy: 0.4860 - 146ms/epoch - 9ms/step\n", "Epoch 59/400\n", "16/16 - 0s - loss: 0.8898 - accuracy: 0.4740 - val_loss: 0.8949 - val_accuracy: 0.4900 - 134ms/epoch - 8ms/step\n", "Epoch 60/400\n", "16/16 - 0s - loss: 0.8859 - accuracy: 0.4740 - val_loss: 0.8907 - val_accuracy: 0.4900 - 142ms/epoch - 9ms/step\n", "Epoch 61/400\n", "16/16 - 0s - loss: 0.8820 - accuracy: 0.4760 - val_loss: 0.8864 - val_accuracy: 0.4880 - 154ms/epoch - 10ms/step\n", "Epoch 62/400\n", "16/16 - 0s - loss: 0.8783 - accuracy: 0.4760 - val_loss: 0.8822 - val_accuracy: 0.4920 - 291ms/epoch - 18ms/step\n", "Epoch 63/400\n", "16/16 - 0s - loss: 0.8747 - accuracy: 0.4760 - val_loss: 0.8778 - val_accuracy: 0.4920 - 135ms/epoch - 8ms/step\n", "Epoch 64/400\n", "16/16 - 0s - loss: 0.8709 - accuracy: 0.4780 - val_loss: 0.8741 - val_accuracy: 0.4920 - 146ms/epoch - 9ms/step\n", "Epoch 65/400\n", "16/16 - 0s - loss: 0.8674 - accuracy: 0.4780 - val_loss: 0.8698 - val_accuracy: 0.4900 - 140ms/epoch - 9ms/step\n", "Epoch 66/400\n", "16/16 - 0s - loss: 0.8638 - accuracy: 0.4800 - val_loss: 0.8659 - val_accuracy: 0.4900 - 100ms/epoch - 6ms/step\n", "Epoch 67/400\n", "16/16 - 0s - loss: 0.8603 - accuracy: 0.4820 - val_loss: 0.8618 - val_accuracy: 0.4920 - 166ms/epoch - 10ms/step\n", "Epoch 68/400\n", "16/16 - 0s - loss: 0.8568 - accuracy: 0.4820 - val_loss: 0.8583 - val_accuracy: 0.4900 - 127ms/epoch - 8ms/step\n", "Epoch 69/400\n", "16/16 - 0s - loss: 0.8533 - accuracy: 0.4820 - val_loss: 0.8543 - val_accuracy: 0.4920 - 113ms/epoch - 7ms/step\n", "Epoch 70/400\n", "16/16 - 0s - loss: 0.8501 - accuracy: 0.4840 - val_loss: 0.8501 - val_accuracy: 0.4920 - 114ms/epoch - 7ms/step\n", "Epoch 71/400\n", "16/16 - 0s - loss: 0.8465 - accuracy: 0.4860 - val_loss: 0.8466 - val_accuracy: 0.4940 - 254ms/epoch - 16ms/step\n", "Epoch 72/400\n", "16/16 - 0s - loss: 0.8433 - accuracy: 0.4840 - val_loss: 0.8426 - val_accuracy: 0.4940 - 303ms/epoch - 19ms/step\n", "Epoch 73/400\n", "16/16 - 0s - loss: 0.8398 - accuracy: 0.4840 - val_loss: 0.8391 - val_accuracy: 0.4940 - 192ms/epoch - 12ms/step\n", "Epoch 74/400\n", "16/16 - 0s - loss: 0.8364 - accuracy: 0.4840 - val_loss: 0.8358 - val_accuracy: 0.4960 - 155ms/epoch - 10ms/step\n", "Epoch 75/400\n", "16/16 - 0s - loss: 0.8332 - accuracy: 0.4840 - val_loss: 0.8318 - val_accuracy: 0.4960 - 138ms/epoch - 9ms/step\n", "Epoch 76/400\n", "16/16 - 0s - loss: 0.8298 - accuracy: 0.4840 - val_loss: 0.8281 - val_accuracy: 0.4980 - 331ms/epoch - 21ms/step\n", "Epoch 77/400\n", "16/16 - 0s - loss: 0.8264 - accuracy: 0.4820 - val_loss: 0.8245 - val_accuracy: 0.5000 - 135ms/epoch - 8ms/step\n", "Epoch 78/400\n", "16/16 - 0s - loss: 0.8233 - accuracy: 0.4880 - val_loss: 0.8206 - val_accuracy: 0.5000 - 160ms/epoch - 10ms/step\n", "Epoch 79/400\n", "16/16 - 0s - loss: 0.8200 - accuracy: 0.4920 - val_loss: 0.8171 - val_accuracy: 0.5020 - 152ms/epoch - 10ms/step\n", "Epoch 80/400\n", "16/16 - 0s - loss: 0.8167 - accuracy: 0.4940 - val_loss: 0.8139 - val_accuracy: 0.5020 - 162ms/epoch - 10ms/step\n", "Epoch 81/400\n", "16/16 - 0s - loss: 0.8135 - accuracy: 0.4940 - val_loss: 0.8101 - val_accuracy: 0.5020 - 109ms/epoch - 7ms/step\n", "Epoch 82/400\n", "16/16 - 0s - loss: 0.8102 - accuracy: 0.4920 - val_loss: 0.8065 - val_accuracy: 0.5080 - 142ms/epoch - 9ms/step\n", "Epoch 83/400\n", "16/16 - 0s - loss: 0.8071 - accuracy: 0.4960 - val_loss: 0.8028 - val_accuracy: 0.5080 - 223ms/epoch - 14ms/step\n", "Epoch 84/400\n", "16/16 - 0s - loss: 0.8037 - accuracy: 0.4960 - val_loss: 0.7993 - val_accuracy: 0.5080 - 174ms/epoch - 11ms/step\n", "Epoch 85/400\n", "16/16 - 0s - loss: 0.8007 - accuracy: 0.4940 - val_loss: 0.7954 - val_accuracy: 0.5100 - 195ms/epoch - 12ms/step\n", "Epoch 86/400\n", "16/16 - 0s - loss: 0.7973 - accuracy: 0.4960 - val_loss: 0.7920 - val_accuracy: 0.5120 - 141ms/epoch - 9ms/step\n", "Epoch 87/400\n", "16/16 - 0s - loss: 0.7942 - accuracy: 0.5000 - val_loss: 0.7885 - val_accuracy: 0.5180 - 195ms/epoch - 12ms/step\n", "Epoch 88/400\n", "16/16 - 0s - loss: 0.7909 - accuracy: 0.4980 - val_loss: 0.7852 - val_accuracy: 0.5220 - 123ms/epoch - 8ms/step\n", "Epoch 89/400\n", "16/16 - 0s - loss: 0.7878 - accuracy: 0.5000 - val_loss: 0.7816 - val_accuracy: 0.5220 - 127ms/epoch - 8ms/step\n", "Epoch 90/400\n", "16/16 - 0s - loss: 0.7847 - accuracy: 0.5020 - val_loss: 0.7778 - val_accuracy: 0.5220 - 258ms/epoch - 16ms/step\n", "Epoch 91/400\n", "16/16 - 0s - loss: 0.7815 - accuracy: 0.5020 - val_loss: 0.7741 - val_accuracy: 0.5220 - 214ms/epoch - 13ms/step\n", "Epoch 92/400\n", "16/16 - 0s - loss: 0.7782 - accuracy: 0.5040 - val_loss: 0.7706 - val_accuracy: 0.5160 - 211ms/epoch - 13ms/step\n", "Epoch 93/400\n", "16/16 - 0s - loss: 0.7750 - accuracy: 0.5020 - val_loss: 0.7671 - val_accuracy: 0.5180 - 136ms/epoch - 9ms/step\n", "Epoch 94/400\n", "16/16 - 0s - loss: 0.7720 - accuracy: 0.5040 - val_loss: 0.7635 - val_accuracy: 0.5220 - 100ms/epoch - 6ms/step\n", "Epoch 95/400\n", "16/16 - 0s - loss: 0.7688 - accuracy: 0.5040 - val_loss: 0.7598 - val_accuracy: 0.5200 - 155ms/epoch - 10ms/step\n", "Epoch 96/400\n", "16/16 - 0s - loss: 0.7656 - accuracy: 0.5040 - val_loss: 0.7563 - val_accuracy: 0.5180 - 170ms/epoch - 11ms/step\n", "Epoch 97/400\n", "16/16 - 0s - loss: 0.7624 - accuracy: 0.5040 - val_loss: 0.7529 - val_accuracy: 0.5180 - 137ms/epoch - 9ms/step\n", "Epoch 98/400\n", "16/16 - 0s - loss: 0.7594 - accuracy: 0.5060 - val_loss: 0.7494 - val_accuracy: 0.5140 - 128ms/epoch - 8ms/step\n", "Epoch 99/400\n", "16/16 - 0s - loss: 0.7562 - accuracy: 0.5080 - val_loss: 0.7458 - val_accuracy: 0.5120 - 163ms/epoch - 10ms/step\n", "Epoch 100/400\n", "16/16 - 0s - loss: 0.7531 - accuracy: 0.5060 - val_loss: 0.7422 - val_accuracy: 0.5120 - 115ms/epoch - 7ms/step\n", "Epoch 101/400\n", "16/16 - 0s - loss: 0.7500 - accuracy: 0.5000 - val_loss: 0.7391 - val_accuracy: 0.5140 - 134ms/epoch - 8ms/step\n", "Epoch 102/400\n", "16/16 - 0s - loss: 0.7469 - accuracy: 0.4960 - val_loss: 0.7356 - val_accuracy: 0.5120 - 209ms/epoch - 13ms/step\n", "Epoch 103/400\n", "16/16 - 0s - loss: 0.7439 - accuracy: 0.4940 - val_loss: 0.7318 - val_accuracy: 0.5140 - 276ms/epoch - 17ms/step\n", "Epoch 104/400\n", "16/16 - 0s - loss: 0.7406 - accuracy: 0.4960 - val_loss: 0.7285 - val_accuracy: 0.5180 - 250ms/epoch - 16ms/step\n", "Epoch 105/400\n", "16/16 - 0s - loss: 0.7376 - accuracy: 0.5020 - val_loss: 0.7249 - val_accuracy: 0.5180 - 145ms/epoch - 9ms/step\n", "Epoch 106/400\n", "16/16 - 0s - loss: 0.7344 - accuracy: 0.5020 - val_loss: 0.7214 - val_accuracy: 0.5220 - 129ms/epoch - 8ms/step\n", "Epoch 107/400\n", "16/16 - 0s - loss: 0.7313 - accuracy: 0.4980 - val_loss: 0.7178 - val_accuracy: 0.5280 - 173ms/epoch - 11ms/step\n", "Epoch 108/400\n", "16/16 - 0s - loss: 0.7282 - accuracy: 0.4980 - val_loss: 0.7145 - val_accuracy: 0.5300 - 223ms/epoch - 14ms/step\n", "Epoch 109/400\n", "16/16 - 0s - loss: 0.7250 - accuracy: 0.4980 - val_loss: 0.7112 - val_accuracy: 0.5320 - 149ms/epoch - 9ms/step\n", "Epoch 110/400\n", "16/16 - 0s - loss: 0.7220 - accuracy: 0.4980 - val_loss: 0.7074 - val_accuracy: 0.5320 - 179ms/epoch - 11ms/step\n", "Epoch 111/400\n", "16/16 - 0s - loss: 0.7189 - accuracy: 0.5020 - val_loss: 0.7038 - val_accuracy: 0.5300 - 275ms/epoch - 17ms/step\n", "Epoch 112/400\n", "16/16 - 0s - loss: 0.7158 - accuracy: 0.5020 - val_loss: 0.7004 - val_accuracy: 0.5300 - 372ms/epoch - 23ms/step\n", "Epoch 113/400\n", "16/16 - 0s - loss: 0.7127 - accuracy: 0.5080 - val_loss: 0.6972 - val_accuracy: 0.5300 - 276ms/epoch - 17ms/step\n", "Epoch 114/400\n", "16/16 - 0s - loss: 0.7097 - accuracy: 0.5080 - val_loss: 0.6938 - val_accuracy: 0.5320 - 186ms/epoch - 12ms/step\n", "Epoch 115/400\n", "16/16 - 0s - loss: 0.7067 - accuracy: 0.5120 - val_loss: 0.6906 - val_accuracy: 0.5340 - 96ms/epoch - 6ms/step\n", "Epoch 116/400\n", "16/16 - 0s - loss: 0.7037 - accuracy: 0.5120 - val_loss: 0.6869 - val_accuracy: 0.5340 - 74ms/epoch - 5ms/step\n", "Epoch 117/400\n", "16/16 - 0s - loss: 0.7006 - accuracy: 0.5120 - val_loss: 0.6834 - val_accuracy: 0.5360 - 73ms/epoch - 5ms/step\n", "Epoch 118/400\n", "16/16 - 0s - loss: 0.6977 - accuracy: 0.5140 - val_loss: 0.6800 - val_accuracy: 0.5360 - 114ms/epoch - 7ms/step\n", "Epoch 119/400\n", "16/16 - 0s - loss: 0.6946 - accuracy: 0.5140 - val_loss: 0.6768 - val_accuracy: 0.5380 - 115ms/epoch - 7ms/step\n", "Epoch 120/400\n", "16/16 - 0s - loss: 0.6917 - accuracy: 0.5160 - val_loss: 0.6736 - val_accuracy: 0.5400 - 76ms/epoch - 5ms/step\n", "Epoch 121/400\n", "16/16 - 0s - loss: 0.6887 - accuracy: 0.5180 - val_loss: 0.6705 - val_accuracy: 0.5420 - 108ms/epoch - 7ms/step\n", "Epoch 122/400\n", "16/16 - 0s - loss: 0.6859 - accuracy: 0.5180 - val_loss: 0.6671 - val_accuracy: 0.5460 - 107ms/epoch - 7ms/step\n", "Epoch 123/400\n", "16/16 - 0s - loss: 0.6829 - accuracy: 0.5220 - val_loss: 0.6642 - val_accuracy: 0.5460 - 68ms/epoch - 4ms/step\n", "Epoch 124/400\n", "16/16 - 0s - loss: 0.6802 - accuracy: 0.5220 - val_loss: 0.6610 - val_accuracy: 0.5480 - 77ms/epoch - 5ms/step\n", "Epoch 125/400\n", "16/16 - 0s - loss: 0.6773 - accuracy: 0.5220 - val_loss: 0.6577 - val_accuracy: 0.5480 - 69ms/epoch - 4ms/step\n", "Epoch 126/400\n", "16/16 - 0s - loss: 0.6744 - accuracy: 0.5220 - val_loss: 0.6548 - val_accuracy: 0.5500 - 71ms/epoch - 4ms/step\n", "Epoch 127/400\n", "16/16 - 0s - loss: 0.6716 - accuracy: 0.5260 - val_loss: 0.6517 - val_accuracy: 0.5500 - 67ms/epoch - 4ms/step\n", "Epoch 128/400\n", "16/16 - 0s - loss: 0.6690 - accuracy: 0.5280 - val_loss: 0.6485 - val_accuracy: 0.5480 - 123ms/epoch - 8ms/step\n", "Epoch 129/400\n", "16/16 - 0s - loss: 0.6662 - accuracy: 0.5320 - val_loss: 0.6456 - val_accuracy: 0.5480 - 72ms/epoch - 4ms/step\n", "Epoch 130/400\n", "16/16 - 0s - loss: 0.6635 - accuracy: 0.5360 - val_loss: 0.6425 - val_accuracy: 0.5500 - 66ms/epoch - 4ms/step\n", "Epoch 131/400\n", "16/16 - 0s - loss: 0.6608 - accuracy: 0.5380 - val_loss: 0.6399 - val_accuracy: 0.5520 - 68ms/epoch - 4ms/step\n", "Epoch 132/400\n", "16/16 - 0s - loss: 0.6582 - accuracy: 0.5360 - val_loss: 0.6367 - val_accuracy: 0.5520 - 66ms/epoch - 4ms/step\n", "Epoch 133/400\n", "16/16 - 0s - loss: 0.6555 - accuracy: 0.5360 - val_loss: 0.6339 - val_accuracy: 0.5520 - 113ms/epoch - 7ms/step\n", "Epoch 134/400\n", "16/16 - 0s - loss: 0.6529 - accuracy: 0.5340 - val_loss: 0.6313 - val_accuracy: 0.5540 - 68ms/epoch - 4ms/step\n", "Epoch 135/400\n", "16/16 - 0s - loss: 0.6504 - accuracy: 0.5380 - val_loss: 0.6285 - val_accuracy: 0.5520 - 107ms/epoch - 7ms/step\n", "Epoch 136/400\n", "16/16 - 0s - loss: 0.6478 - accuracy: 0.5420 - val_loss: 0.6260 - val_accuracy: 0.5580 - 108ms/epoch - 7ms/step\n", "Epoch 137/400\n", "16/16 - 0s - loss: 0.6454 - accuracy: 0.5540 - val_loss: 0.6231 - val_accuracy: 0.5880 - 108ms/epoch - 7ms/step\n", "Epoch 138/400\n", "16/16 - 0s - loss: 0.6429 - accuracy: 0.5780 - val_loss: 0.6206 - val_accuracy: 0.6480 - 69ms/epoch - 4ms/step\n", "Epoch 139/400\n", "16/16 - 0s - loss: 0.6405 - accuracy: 0.6240 - val_loss: 0.6180 - val_accuracy: 0.6940 - 65ms/epoch - 4ms/step\n", "Epoch 140/400\n", "16/16 - 0s - loss: 0.6379 - accuracy: 0.6320 - val_loss: 0.6156 - val_accuracy: 0.7020 - 73ms/epoch - 5ms/step\n", "Epoch 141/400\n", "16/16 - 0s - loss: 0.6357 - accuracy: 0.6540 - val_loss: 0.6130 - val_accuracy: 0.7220 - 65ms/epoch - 4ms/step\n", "Epoch 142/400\n", "16/16 - 0s - loss: 0.6333 - accuracy: 0.6640 - val_loss: 0.6105 - val_accuracy: 0.7340 - 70ms/epoch - 4ms/step\n", "Epoch 143/400\n", "16/16 - 0s - loss: 0.6311 - accuracy: 0.6760 - val_loss: 0.6082 - val_accuracy: 0.7380 - 105ms/epoch - 7ms/step\n", "Epoch 144/400\n", "16/16 - 0s - loss: 0.6287 - accuracy: 0.6780 - val_loss: 0.6061 - val_accuracy: 0.7480 - 71ms/epoch - 4ms/step\n", "Epoch 145/400\n", "16/16 - 0s - loss: 0.6266 - accuracy: 0.6860 - val_loss: 0.6035 - val_accuracy: 0.7500 - 68ms/epoch - 4ms/step\n", "Epoch 146/400\n", "16/16 - 0s - loss: 0.6243 - accuracy: 0.6980 - val_loss: 0.6013 - val_accuracy: 0.7540 - 69ms/epoch - 4ms/step\n", "Epoch 147/400\n", "16/16 - 0s - loss: 0.6221 - accuracy: 0.7000 - val_loss: 0.5990 - val_accuracy: 0.7500 - 66ms/epoch - 4ms/step\n", "Epoch 148/400\n", "16/16 - 0s - loss: 0.6200 - accuracy: 0.7020 - val_loss: 0.5969 - val_accuracy: 0.7520 - 70ms/epoch - 4ms/step\n", "Epoch 149/400\n", "16/16 - 0s - loss: 0.6179 - accuracy: 0.7060 - val_loss: 0.5946 - val_accuracy: 0.7500 - 79ms/epoch - 5ms/step\n", "Epoch 150/400\n", "16/16 - 0s - loss: 0.6159 - accuracy: 0.7080 - val_loss: 0.5925 - val_accuracy: 0.7540 - 64ms/epoch - 4ms/step\n", "Epoch 151/400\n", "16/16 - 0s - loss: 0.6138 - accuracy: 0.7120 - val_loss: 0.5904 - val_accuracy: 0.7560 - 64ms/epoch - 4ms/step\n", "Epoch 152/400\n", "16/16 - 0s - loss: 0.6118 - accuracy: 0.7220 - val_loss: 0.5883 - val_accuracy: 0.7580 - 70ms/epoch - 4ms/step\n", "Epoch 153/400\n", "16/16 - 0s - loss: 0.6100 - accuracy: 0.7200 - val_loss: 0.5865 - val_accuracy: 0.7620 - 114ms/epoch - 7ms/step\n", "Epoch 154/400\n", "16/16 - 0s - loss: 0.6080 - accuracy: 0.7140 - val_loss: 0.5843 - val_accuracy: 0.7580 - 109ms/epoch - 7ms/step\n", "Epoch 155/400\n", "16/16 - 0s - loss: 0.6061 - accuracy: 0.7160 - val_loss: 0.5825 - val_accuracy: 0.7600 - 74ms/epoch - 5ms/step\n", "Epoch 156/400\n", "16/16 - 0s - loss: 0.6041 - accuracy: 0.7160 - val_loss: 0.5807 - val_accuracy: 0.7580 - 69ms/epoch - 4ms/step\n", "Epoch 157/400\n", "16/16 - 0s - loss: 0.6023 - accuracy: 0.7160 - val_loss: 0.5787 - val_accuracy: 0.7540 - 67ms/epoch - 4ms/step\n", "Epoch 158/400\n", "16/16 - 0s - loss: 0.6005 - accuracy: 0.7160 - val_loss: 0.5769 - val_accuracy: 0.7540 - 68ms/epoch - 4ms/step\n", "Epoch 159/400\n", "16/16 - 0s - loss: 0.5987 - accuracy: 0.7160 - val_loss: 0.5751 - val_accuracy: 0.7600 - 64ms/epoch - 4ms/step\n", "Epoch 160/400\n", "16/16 - 0s - loss: 0.5970 - accuracy: 0.7100 - val_loss: 0.5734 - val_accuracy: 0.7600 - 107ms/epoch - 7ms/step\n", "Epoch 161/400\n", "16/16 - 0s - loss: 0.5952 - accuracy: 0.7080 - val_loss: 0.5717 - val_accuracy: 0.7620 - 68ms/epoch - 4ms/step\n", "Epoch 162/400\n", "16/16 - 0s - loss: 0.5936 - accuracy: 0.7000 - val_loss: 0.5699 - val_accuracy: 0.7620 - 69ms/epoch - 4ms/step\n", "Epoch 163/400\n", "16/16 - 0s - loss: 0.5920 - accuracy: 0.7060 - val_loss: 0.5683 - val_accuracy: 0.7660 - 69ms/epoch - 4ms/step\n", "Epoch 164/400\n", "16/16 - 0s - loss: 0.5902 - accuracy: 0.7100 - val_loss: 0.5667 - val_accuracy: 0.7660 - 104ms/epoch - 6ms/step\n", "Epoch 165/400\n", "16/16 - 0s - loss: 0.5887 - accuracy: 0.7100 - val_loss: 0.5651 - val_accuracy: 0.7680 - 77ms/epoch - 5ms/step\n", "Epoch 166/400\n", "16/16 - 0s - loss: 0.5870 - accuracy: 0.7100 - val_loss: 0.5636 - val_accuracy: 0.7700 - 64ms/epoch - 4ms/step\n", "Epoch 167/400\n", "16/16 - 0s - loss: 0.5856 - accuracy: 0.7060 - val_loss: 0.5618 - val_accuracy: 0.7640 - 104ms/epoch - 6ms/step\n", "Epoch 168/400\n", "16/16 - 0s - loss: 0.5840 - accuracy: 0.7140 - val_loss: 0.5604 - val_accuracy: 0.7620 - 64ms/epoch - 4ms/step\n", "Epoch 169/400\n", "16/16 - 0s - loss: 0.5824 - accuracy: 0.7140 - val_loss: 0.5589 - val_accuracy: 0.7620 - 64ms/epoch - 4ms/step\n", "Epoch 170/400\n", "16/16 - 0s - loss: 0.5809 - accuracy: 0.7160 - val_loss: 0.5574 - val_accuracy: 0.7660 - 67ms/epoch - 4ms/step\n", "Epoch 171/400\n", "16/16 - 0s - loss: 0.5795 - accuracy: 0.7180 - val_loss: 0.5560 - val_accuracy: 0.7640 - 64ms/epoch - 4ms/step\n", "Epoch 172/400\n", "16/16 - 0s - loss: 0.5779 - accuracy: 0.7220 - val_loss: 0.5545 - val_accuracy: 0.7660 - 105ms/epoch - 7ms/step\n", "Epoch 173/400\n", "16/16 - 0s - loss: 0.5766 - accuracy: 0.7260 - val_loss: 0.5530 - val_accuracy: 0.7620 - 106ms/epoch - 7ms/step\n", "Epoch 174/400\n", "16/16 - 0s - loss: 0.5751 - accuracy: 0.7260 - val_loss: 0.5515 - val_accuracy: 0.7580 - 73ms/epoch - 5ms/step\n", "Epoch 175/400\n", "16/16 - 0s - loss: 0.5736 - accuracy: 0.7260 - val_loss: 0.5503 - val_accuracy: 0.7620 - 67ms/epoch - 4ms/step\n", "Epoch 176/400\n", "16/16 - 0s - loss: 0.5724 - accuracy: 0.7280 - val_loss: 0.5489 - val_accuracy: 0.7640 - 64ms/epoch - 4ms/step\n", "Epoch 177/400\n", "16/16 - 0s - loss: 0.5709 - accuracy: 0.7320 - val_loss: 0.5476 - val_accuracy: 0.7640 - 113ms/epoch - 7ms/step\n", "Epoch 178/400\n", "16/16 - 0s - loss: 0.5696 - accuracy: 0.7360 - val_loss: 0.5462 - val_accuracy: 0.7660 - 77ms/epoch - 5ms/step\n", "Epoch 179/400\n", "16/16 - 0s - loss: 0.5683 - accuracy: 0.7360 - val_loss: 0.5448 - val_accuracy: 0.7620 - 64ms/epoch - 4ms/step\n", "Epoch 180/400\n", "16/16 - 0s - loss: 0.5669 - accuracy: 0.7380 - val_loss: 0.5435 - val_accuracy: 0.7620 - 68ms/epoch - 4ms/step\n", "Epoch 181/400\n", "16/16 - 0s - loss: 0.5657 - accuracy: 0.7380 - val_loss: 0.5421 - val_accuracy: 0.7640 - 68ms/epoch - 4ms/step\n", "Epoch 182/400\n", "16/16 - 0s - loss: 0.5644 - accuracy: 0.7340 - val_loss: 0.5407 - val_accuracy: 0.7640 - 66ms/epoch - 4ms/step\n", "Epoch 183/400\n", "16/16 - 0s - loss: 0.5631 - accuracy: 0.7360 - val_loss: 0.5395 - val_accuracy: 0.7620 - 65ms/epoch - 4ms/step\n", "Epoch 184/400\n", "16/16 - 0s - loss: 0.5618 - accuracy: 0.7400 - val_loss: 0.5383 - val_accuracy: 0.7620 - 63ms/epoch - 4ms/step\n", "Epoch 185/400\n", "16/16 - 0s - loss: 0.5606 - accuracy: 0.7480 - val_loss: 0.5372 - val_accuracy: 0.7640 - 106ms/epoch - 7ms/step\n", "Epoch 186/400\n", "16/16 - 0s - loss: 0.5593 - accuracy: 0.7480 - val_loss: 0.5360 - val_accuracy: 0.7680 - 68ms/epoch - 4ms/step\n", "Epoch 187/400\n", "16/16 - 0s - loss: 0.5581 - accuracy: 0.7480 - val_loss: 0.5347 - val_accuracy: 0.7700 - 71ms/epoch - 4ms/step\n", "Epoch 188/400\n", "16/16 - 0s - loss: 0.5570 - accuracy: 0.7480 - val_loss: 0.5335 - val_accuracy: 0.7740 - 111ms/epoch - 7ms/step\n", "Epoch 189/400\n", "16/16 - 0s - loss: 0.5558 - accuracy: 0.7500 - val_loss: 0.5323 - val_accuracy: 0.7720 - 66ms/epoch - 4ms/step\n", "Epoch 190/400\n", "16/16 - 0s - loss: 0.5546 - accuracy: 0.7500 - val_loss: 0.5311 - val_accuracy: 0.7720 - 114ms/epoch - 7ms/step\n", "Epoch 191/400\n", "16/16 - 0s - loss: 0.5534 - accuracy: 0.7540 - val_loss: 0.5300 - val_accuracy: 0.7800 - 70ms/epoch - 4ms/step\n", "Epoch 192/400\n", "16/16 - 0s - loss: 0.5523 - accuracy: 0.7580 - val_loss: 0.5288 - val_accuracy: 0.7800 - 106ms/epoch - 7ms/step\n", "Epoch 193/400\n", "16/16 - 0s - loss: 0.5512 - accuracy: 0.7620 - val_loss: 0.5278 - val_accuracy: 0.7820 - 70ms/epoch - 4ms/step\n", "Epoch 194/400\n", "16/16 - 0s - loss: 0.5502 - accuracy: 0.7620 - val_loss: 0.5266 - val_accuracy: 0.7800 - 64ms/epoch - 4ms/step\n", "Epoch 195/400\n", "16/16 - 0s - loss: 0.5491 - accuracy: 0.7620 - val_loss: 0.5255 - val_accuracy: 0.7880 - 108ms/epoch - 7ms/step\n", "Epoch 196/400\n", "16/16 - 0s - loss: 0.5480 - accuracy: 0.7620 - val_loss: 0.5244 - val_accuracy: 0.7900 - 65ms/epoch - 4ms/step\n", "Epoch 197/400\n", "16/16 - 0s - loss: 0.5470 - accuracy: 0.7620 - val_loss: 0.5234 - val_accuracy: 0.7900 - 73ms/epoch - 5ms/step\n", "Epoch 198/400\n", "16/16 - 0s - loss: 0.5459 - accuracy: 0.7680 - val_loss: 0.5224 - val_accuracy: 0.7900 - 69ms/epoch - 4ms/step\n", "Epoch 199/400\n", "16/16 - 0s - loss: 0.5449 - accuracy: 0.7680 - val_loss: 0.5212 - val_accuracy: 0.7860 - 73ms/epoch - 5ms/step\n", "Epoch 200/400\n", "16/16 - 0s - loss: 0.5439 - accuracy: 0.7660 - val_loss: 0.5201 - val_accuracy: 0.7900 - 74ms/epoch - 5ms/step\n", "Epoch 201/400\n", "16/16 - 0s - loss: 0.5430 - accuracy: 0.7680 - val_loss: 0.5193 - val_accuracy: 0.7900 - 69ms/epoch - 4ms/step\n", "Epoch 202/400\n", "16/16 - 0s - loss: 0.5420 - accuracy: 0.7740 - val_loss: 0.5183 - val_accuracy: 0.7900 - 121ms/epoch - 8ms/step\n", "Epoch 203/400\n", "16/16 - 0s - loss: 0.5410 - accuracy: 0.7760 - val_loss: 0.5172 - val_accuracy: 0.7900 - 68ms/epoch - 4ms/step\n", "Epoch 204/400\n", "16/16 - 0s - loss: 0.5400 - accuracy: 0.7760 - val_loss: 0.5162 - val_accuracy: 0.7920 - 113ms/epoch - 7ms/step\n", "Epoch 205/400\n", "16/16 - 0s - loss: 0.5391 - accuracy: 0.7780 - val_loss: 0.5152 - val_accuracy: 0.7960 - 72ms/epoch - 5ms/step\n", "Epoch 206/400\n", "16/16 - 0s - loss: 0.5382 - accuracy: 0.7840 - val_loss: 0.5143 - val_accuracy: 0.7980 - 70ms/epoch - 4ms/step\n", "Epoch 207/400\n", "16/16 - 0s - loss: 0.5373 - accuracy: 0.7860 - val_loss: 0.5133 - val_accuracy: 0.7980 - 66ms/epoch - 4ms/step\n", "Epoch 208/400\n", "16/16 - 0s - loss: 0.5364 - accuracy: 0.7860 - val_loss: 0.5125 - val_accuracy: 0.7980 - 72ms/epoch - 4ms/step\n", "Epoch 209/400\n", "16/16 - 0s - loss: 0.5355 - accuracy: 0.7880 - val_loss: 0.5114 - val_accuracy: 0.7980 - 71ms/epoch - 4ms/step\n", "Epoch 210/400\n", "16/16 - 0s - loss: 0.5346 - accuracy: 0.7880 - val_loss: 0.5105 - val_accuracy: 0.7980 - 74ms/epoch - 5ms/step\n", "Epoch 211/400\n", "16/16 - 0s - loss: 0.5337 - accuracy: 0.7940 - val_loss: 0.5097 - val_accuracy: 0.8000 - 69ms/epoch - 4ms/step\n", "Epoch 212/400\n", "16/16 - 0s - loss: 0.5329 - accuracy: 0.7940 - val_loss: 0.5087 - val_accuracy: 0.8000 - 109ms/epoch - 7ms/step\n", "Epoch 213/400\n", "16/16 - 0s - loss: 0.5320 - accuracy: 0.7960 - val_loss: 0.5079 - val_accuracy: 0.8000 - 106ms/epoch - 7ms/step\n", "Epoch 214/400\n", "16/16 - 0s - loss: 0.5312 - accuracy: 0.8020 - val_loss: 0.5071 - val_accuracy: 0.7960 - 72ms/epoch - 5ms/step\n", "Epoch 215/400\n", "16/16 - 0s - loss: 0.5304 - accuracy: 0.8000 - val_loss: 0.5061 - val_accuracy: 0.7960 - 65ms/epoch - 4ms/step\n", "Epoch 216/400\n", "16/16 - 0s - loss: 0.5296 - accuracy: 0.7980 - val_loss: 0.5052 - val_accuracy: 0.7960 - 69ms/epoch - 4ms/step\n", "Epoch 217/400\n", "16/16 - 0s - loss: 0.5288 - accuracy: 0.7960 - val_loss: 0.5042 - val_accuracy: 0.7980 - 64ms/epoch - 4ms/step\n", "Epoch 218/400\n", "16/16 - 0s - loss: 0.5279 - accuracy: 0.7980 - val_loss: 0.5034 - val_accuracy: 0.7980 - 71ms/epoch - 4ms/step\n", "Epoch 219/400\n", "16/16 - 0s - loss: 0.5272 - accuracy: 0.8000 - val_loss: 0.5027 - val_accuracy: 0.8020 - 66ms/epoch - 4ms/step\n", "Epoch 220/400\n", "16/16 - 0s - loss: 0.5265 - accuracy: 0.8000 - val_loss: 0.5019 - val_accuracy: 0.8000 - 65ms/epoch - 4ms/step\n", "Epoch 221/400\n", "16/16 - 0s - loss: 0.5257 - accuracy: 0.8000 - val_loss: 0.5010 - val_accuracy: 0.8000 - 69ms/epoch - 4ms/step\n", "Epoch 222/400\n", "16/16 - 0s - loss: 0.5249 - accuracy: 0.8000 - val_loss: 0.5002 - val_accuracy: 0.7980 - 65ms/epoch - 4ms/step\n", "Epoch 223/400\n", "16/16 - 0s - loss: 0.5242 - accuracy: 0.8020 - val_loss: 0.4995 - val_accuracy: 0.8020 - 68ms/epoch - 4ms/step\n", "Epoch 224/400\n", "16/16 - 0s - loss: 0.5235 - accuracy: 0.8040 - val_loss: 0.4987 - val_accuracy: 0.8020 - 66ms/epoch - 4ms/step\n", "Epoch 225/400\n", "16/16 - 0s - loss: 0.5228 - accuracy: 0.8040 - val_loss: 0.4978 - val_accuracy: 0.8020 - 108ms/epoch - 7ms/step\n", "Epoch 226/400\n", "16/16 - 0s - loss: 0.5222 - accuracy: 0.8020 - val_loss: 0.4969 - val_accuracy: 0.7980 - 113ms/epoch - 7ms/step\n", "Epoch 227/400\n", "16/16 - 0s - loss: 0.5214 - accuracy: 0.8040 - val_loss: 0.4962 - val_accuracy: 0.8020 - 68ms/epoch - 4ms/step\n", "Epoch 228/400\n", "16/16 - 0s - loss: 0.5207 - accuracy: 0.8060 - val_loss: 0.4955 - val_accuracy: 0.8020 - 69ms/epoch - 4ms/step\n", "Epoch 229/400\n", "16/16 - 0s - loss: 0.5201 - accuracy: 0.8020 - val_loss: 0.4947 - val_accuracy: 0.8020 - 108ms/epoch - 7ms/step\n", "Epoch 230/400\n", "16/16 - 0s - loss: 0.5194 - accuracy: 0.8060 - val_loss: 0.4940 - val_accuracy: 0.8020 - 67ms/epoch - 4ms/step\n", "Epoch 231/400\n", "16/16 - 0s - loss: 0.5187 - accuracy: 0.8040 - val_loss: 0.4933 - val_accuracy: 0.8020 - 67ms/epoch - 4ms/step\n", "Epoch 232/400\n", "16/16 - 0s - loss: 0.5181 - accuracy: 0.8020 - val_loss: 0.4926 - val_accuracy: 0.8020 - 68ms/epoch - 4ms/step\n", "Epoch 233/400\n", "16/16 - 0s - loss: 0.5174 - accuracy: 0.8020 - val_loss: 0.4918 - val_accuracy: 0.8020 - 66ms/epoch - 4ms/step\n", "Epoch 234/400\n", "16/16 - 0s - loss: 0.5168 - accuracy: 0.8020 - val_loss: 0.4911 - val_accuracy: 0.8020 - 64ms/epoch - 4ms/step\n", "Epoch 235/400\n", "16/16 - 0s - loss: 0.5161 - accuracy: 0.8020 - val_loss: 0.4903 - val_accuracy: 0.8020 - 67ms/epoch - 4ms/step\n", "Epoch 236/400\n", "16/16 - 0s - loss: 0.5155 - accuracy: 0.8020 - val_loss: 0.4897 - val_accuracy: 0.8020 - 71ms/epoch - 4ms/step\n", "Epoch 237/400\n", "16/16 - 0s - loss: 0.5149 - accuracy: 0.8000 - val_loss: 0.4890 - val_accuracy: 0.8020 - 65ms/epoch - 4ms/step\n", "Epoch 238/400\n", "16/16 - 0s - loss: 0.5143 - accuracy: 0.7980 - val_loss: 0.4883 - val_accuracy: 0.8060 - 65ms/epoch - 4ms/step\n", "Epoch 239/400\n", "16/16 - 0s - loss: 0.5136 - accuracy: 0.7980 - val_loss: 0.4877 - val_accuracy: 0.8060 - 105ms/epoch - 7ms/step\n", "Epoch 240/400\n", "16/16 - 0s - loss: 0.5131 - accuracy: 0.8000 - val_loss: 0.4869 - val_accuracy: 0.8060 - 113ms/epoch - 7ms/step\n", "Epoch 241/400\n", "16/16 - 0s - loss: 0.5124 - accuracy: 0.7960 - val_loss: 0.4863 - val_accuracy: 0.8060 - 66ms/epoch - 4ms/step\n", "Epoch 242/400\n", "16/16 - 0s - loss: 0.5119 - accuracy: 0.7980 - val_loss: 0.4855 - val_accuracy: 0.8060 - 74ms/epoch - 5ms/step\n", "Epoch 243/400\n", "16/16 - 0s - loss: 0.5113 - accuracy: 0.7960 - val_loss: 0.4849 - val_accuracy: 0.8060 - 70ms/epoch - 4ms/step\n", "Epoch 244/400\n", "16/16 - 0s - loss: 0.5107 - accuracy: 0.7980 - val_loss: 0.4842 - val_accuracy: 0.8060 - 65ms/epoch - 4ms/step\n", "Epoch 245/400\n", "16/16 - 0s - loss: 0.5101 - accuracy: 0.7980 - val_loss: 0.4836 - val_accuracy: 0.8040 - 64ms/epoch - 4ms/step\n", "Epoch 246/400\n", "16/16 - 0s - loss: 0.5096 - accuracy: 0.7980 - val_loss: 0.4829 - val_accuracy: 0.8040 - 69ms/epoch - 4ms/step\n", "Epoch 247/400\n", "16/16 - 0s - loss: 0.5090 - accuracy: 0.8000 - val_loss: 0.4823 - val_accuracy: 0.8040 - 67ms/epoch - 4ms/step\n", "Epoch 248/400\n", "16/16 - 0s - loss: 0.5085 - accuracy: 0.8020 - val_loss: 0.4817 - val_accuracy: 0.8040 - 108ms/epoch - 7ms/step\n", "Epoch 249/400\n", "16/16 - 0s - loss: 0.5080 - accuracy: 0.8000 - val_loss: 0.4811 - val_accuracy: 0.8040 - 69ms/epoch - 4ms/step\n", "Epoch 250/400\n", "16/16 - 0s - loss: 0.5074 - accuracy: 0.8000 - val_loss: 0.4803 - val_accuracy: 0.8040 - 69ms/epoch - 4ms/step\n", "Epoch 251/400\n", "16/16 - 0s - loss: 0.5069 - accuracy: 0.7980 - val_loss: 0.4796 - val_accuracy: 0.8060 - 74ms/epoch - 5ms/step\n", "Epoch 252/400\n", "16/16 - 0s - loss: 0.5064 - accuracy: 0.8020 - val_loss: 0.4789 - val_accuracy: 0.8060 - 68ms/epoch - 4ms/step\n", "Epoch 253/400\n", "16/16 - 0s - loss: 0.5059 - accuracy: 0.8000 - val_loss: 0.4784 - val_accuracy: 0.8060 - 107ms/epoch - 7ms/step\n", "Epoch 254/400\n", "16/16 - 0s - loss: 0.5054 - accuracy: 0.8000 - val_loss: 0.4778 - val_accuracy: 0.8060 - 68ms/epoch - 4ms/step\n", "Epoch 255/400\n", "16/16 - 0s - loss: 0.5049 - accuracy: 0.8020 - val_loss: 0.4772 - val_accuracy: 0.8080 - 66ms/epoch - 4ms/step\n", "Epoch 256/400\n", "16/16 - 0s - loss: 0.5044 - accuracy: 0.8020 - val_loss: 0.4766 - val_accuracy: 0.8080 - 107ms/epoch - 7ms/step\n", "Epoch 257/400\n", "16/16 - 0s - loss: 0.5038 - accuracy: 0.8040 - val_loss: 0.4760 - val_accuracy: 0.8080 - 68ms/epoch - 4ms/step\n", "Epoch 258/400\n", "16/16 - 0s - loss: 0.5034 - accuracy: 0.8040 - val_loss: 0.4754 - val_accuracy: 0.8140 - 69ms/epoch - 4ms/step\n", "Epoch 259/400\n", "16/16 - 0s - loss: 0.5029 - accuracy: 0.8060 - val_loss: 0.4748 - val_accuracy: 0.8140 - 69ms/epoch - 4ms/step\n", "Epoch 260/400\n", "16/16 - 0s - loss: 0.5024 - accuracy: 0.8060 - val_loss: 0.4741 - val_accuracy: 0.8140 - 65ms/epoch - 4ms/step\n", "Epoch 261/400\n", "16/16 - 0s - loss: 0.5020 - accuracy: 0.8080 - val_loss: 0.4737 - val_accuracy: 0.8140 - 69ms/epoch - 4ms/step\n", "Epoch 262/400\n", "16/16 - 0s - loss: 0.5016 - accuracy: 0.8100 - val_loss: 0.4732 - val_accuracy: 0.8140 - 107ms/epoch - 7ms/step\n", "Epoch 263/400\n", "16/16 - 0s - loss: 0.5010 - accuracy: 0.8100 - val_loss: 0.4726 - val_accuracy: 0.8120 - 70ms/epoch - 4ms/step\n", "Epoch 264/400\n", "16/16 - 0s - loss: 0.5006 - accuracy: 0.8100 - val_loss: 0.4719 - val_accuracy: 0.8100 - 125ms/epoch - 8ms/step\n", "Epoch 265/400\n", "16/16 - 0s - loss: 0.5001 - accuracy: 0.8060 - val_loss: 0.4713 - val_accuracy: 0.8120 - 71ms/epoch - 4ms/step\n", "Epoch 266/400\n", "16/16 - 0s - loss: 0.4998 - accuracy: 0.8100 - val_loss: 0.4708 - val_accuracy: 0.8100 - 68ms/epoch - 4ms/step\n", "Epoch 267/400\n", "16/16 - 0s - loss: 0.4992 - accuracy: 0.8100 - val_loss: 0.4703 - val_accuracy: 0.8100 - 105ms/epoch - 7ms/step\n", "Epoch 268/400\n", "16/16 - 0s - loss: 0.4989 - accuracy: 0.8100 - val_loss: 0.4697 - val_accuracy: 0.8100 - 68ms/epoch - 4ms/step\n", "Epoch 269/400\n", "16/16 - 0s - loss: 0.4984 - accuracy: 0.8080 - val_loss: 0.4691 - val_accuracy: 0.8100 - 65ms/epoch - 4ms/step\n", "Epoch 270/400\n", "16/16 - 0s - loss: 0.4980 - accuracy: 0.8100 - val_loss: 0.4687 - val_accuracy: 0.8120 - 72ms/epoch - 5ms/step\n", "Epoch 271/400\n", "16/16 - 0s - loss: 0.4976 - accuracy: 0.8100 - val_loss: 0.4682 - val_accuracy: 0.8120 - 71ms/epoch - 4ms/step\n", "Epoch 272/400\n", "16/16 - 0s - loss: 0.4972 - accuracy: 0.8100 - val_loss: 0.4676 - val_accuracy: 0.8100 - 71ms/epoch - 4ms/step\n", "Epoch 273/400\n", "16/16 - 0s - loss: 0.4967 - accuracy: 0.8120 - val_loss: 0.4672 - val_accuracy: 0.8120 - 69ms/epoch - 4ms/step\n", "Epoch 274/400\n", "16/16 - 0s - loss: 0.4964 - accuracy: 0.8140 - val_loss: 0.4668 - val_accuracy: 0.8140 - 67ms/epoch - 4ms/step\n", "Epoch 275/400\n", "16/16 - 0s - loss: 0.4959 - accuracy: 0.8140 - val_loss: 0.4663 - val_accuracy: 0.8120 - 66ms/epoch - 4ms/step\n", "Epoch 276/400\n", "16/16 - 0s - loss: 0.4956 - accuracy: 0.8140 - val_loss: 0.4658 - val_accuracy: 0.8140 - 69ms/epoch - 4ms/step\n", "Epoch 277/400\n", "16/16 - 0s - loss: 0.4951 - accuracy: 0.8140 - val_loss: 0.4652 - val_accuracy: 0.8120 - 75ms/epoch - 5ms/step\n", "Epoch 278/400\n", "16/16 - 0s - loss: 0.4948 - accuracy: 0.8140 - val_loss: 0.4647 - val_accuracy: 0.8120 - 109ms/epoch - 7ms/step\n", "Epoch 279/400\n", "16/16 - 0s - loss: 0.4943 - accuracy: 0.8120 - val_loss: 0.4642 - val_accuracy: 0.8120 - 65ms/epoch - 4ms/step\n", "Epoch 280/400\n", "16/16 - 0s - loss: 0.4940 - accuracy: 0.8120 - val_loss: 0.4636 - val_accuracy: 0.8120 - 67ms/epoch - 4ms/step\n", "Epoch 281/400\n", "16/16 - 0s - loss: 0.4936 - accuracy: 0.8100 - val_loss: 0.4631 - val_accuracy: 0.8140 - 65ms/epoch - 4ms/step\n", "Epoch 282/400\n", "16/16 - 0s - loss: 0.4931 - accuracy: 0.8100 - val_loss: 0.4628 - val_accuracy: 0.8120 - 68ms/epoch - 4ms/step\n", "Epoch 283/400\n", "16/16 - 0s - loss: 0.4928 - accuracy: 0.8100 - val_loss: 0.4622 - val_accuracy: 0.8140 - 66ms/epoch - 4ms/step\n", "Epoch 284/400\n", "16/16 - 0s - loss: 0.4923 - accuracy: 0.8100 - val_loss: 0.4617 - val_accuracy: 0.8140 - 69ms/epoch - 4ms/step\n", "Epoch 285/400\n", "16/16 - 0s - loss: 0.4920 - accuracy: 0.8060 - val_loss: 0.4612 - val_accuracy: 0.8160 - 109ms/epoch - 7ms/step\n", "Epoch 286/400\n", "16/16 - 0s - loss: 0.4915 - accuracy: 0.8080 - val_loss: 0.4607 - val_accuracy: 0.8140 - 70ms/epoch - 4ms/step\n", "Epoch 287/400\n", "16/16 - 0s - loss: 0.4911 - accuracy: 0.8080 - val_loss: 0.4602 - val_accuracy: 0.8140 - 111ms/epoch - 7ms/step\n", "Epoch 288/400\n", "16/16 - 0s - loss: 0.4907 - accuracy: 0.8080 - val_loss: 0.4596 - val_accuracy: 0.8140 - 69ms/epoch - 4ms/step\n", "Epoch 289/400\n", "16/16 - 0s - loss: 0.4903 - accuracy: 0.8080 - val_loss: 0.4591 - val_accuracy: 0.8140 - 78ms/epoch - 5ms/step\n", "Epoch 290/400\n", "16/16 - 0s - loss: 0.4899 - accuracy: 0.8060 - val_loss: 0.4585 - val_accuracy: 0.8120 - 68ms/epoch - 4ms/step\n", "Epoch 291/400\n", "16/16 - 0s - loss: 0.4895 - accuracy: 0.8080 - val_loss: 0.4581 - val_accuracy: 0.8140 - 68ms/epoch - 4ms/step\n", "Epoch 292/400\n", "16/16 - 0s - loss: 0.4891 - accuracy: 0.8060 - val_loss: 0.4575 - val_accuracy: 0.8160 - 109ms/epoch - 7ms/step\n", "Epoch 293/400\n", "16/16 - 0s - loss: 0.4887 - accuracy: 0.8080 - val_loss: 0.4570 - val_accuracy: 0.8160 - 64ms/epoch - 4ms/step\n", "Epoch 294/400\n", "16/16 - 0s - loss: 0.4882 - accuracy: 0.8060 - val_loss: 0.4564 - val_accuracy: 0.8140 - 73ms/epoch - 5ms/step\n", "Epoch 295/400\n", "16/16 - 0s - loss: 0.4878 - accuracy: 0.8060 - val_loss: 0.4559 - val_accuracy: 0.8140 - 75ms/epoch - 5ms/step\n", "Epoch 296/400\n", "16/16 - 0s - loss: 0.4873 - accuracy: 0.8040 - val_loss: 0.4553 - val_accuracy: 0.8140 - 114ms/epoch - 7ms/step\n", "Epoch 297/400\n", "16/16 - 0s - loss: 0.4870 - accuracy: 0.8060 - val_loss: 0.4549 - val_accuracy: 0.8160 - 114ms/epoch - 7ms/step\n", "Epoch 298/400\n", "16/16 - 0s - loss: 0.4864 - accuracy: 0.8080 - val_loss: 0.4543 - val_accuracy: 0.8140 - 68ms/epoch - 4ms/step\n", "Epoch 299/400\n", "16/16 - 0s - loss: 0.4860 - accuracy: 0.8060 - val_loss: 0.4537 - val_accuracy: 0.8140 - 76ms/epoch - 5ms/step\n", "Epoch 300/400\n", "16/16 - 0s - loss: 0.4855 - accuracy: 0.8040 - val_loss: 0.4531 - val_accuracy: 0.8120 - 65ms/epoch - 4ms/step\n", "Epoch 301/400\n", "16/16 - 0s - loss: 0.4850 - accuracy: 0.8040 - val_loss: 0.4525 - val_accuracy: 0.8120 - 78ms/epoch - 5ms/step\n", "Epoch 302/400\n", "16/16 - 0s - loss: 0.4845 - accuracy: 0.8020 - val_loss: 0.4519 - val_accuracy: 0.8120 - 66ms/epoch - 4ms/step\n", "Epoch 303/400\n", "16/16 - 0s - loss: 0.4841 - accuracy: 0.8020 - val_loss: 0.4514 - val_accuracy: 0.8140 - 65ms/epoch - 4ms/step\n", "Epoch 304/400\n", "16/16 - 0s - loss: 0.4837 - accuracy: 0.8020 - val_loss: 0.4507 - val_accuracy: 0.8120 - 68ms/epoch - 4ms/step\n", "Epoch 305/400\n", "16/16 - 0s - loss: 0.4831 - accuracy: 0.8020 - val_loss: 0.4503 - val_accuracy: 0.8140 - 107ms/epoch - 7ms/step\n", "Epoch 306/400\n", "16/16 - 0s - loss: 0.4826 - accuracy: 0.8020 - val_loss: 0.4496 - val_accuracy: 0.8140 - 72ms/epoch - 4ms/step\n", "Epoch 307/400\n", "16/16 - 0s - loss: 0.4822 - accuracy: 0.8020 - val_loss: 0.4490 - val_accuracy: 0.8100 - 108ms/epoch - 7ms/step\n", "Epoch 308/400\n", "16/16 - 0s - loss: 0.4817 - accuracy: 0.8020 - val_loss: 0.4486 - val_accuracy: 0.8140 - 109ms/epoch - 7ms/step\n", "Epoch 309/400\n", "16/16 - 0s - loss: 0.4811 - accuracy: 0.8020 - val_loss: 0.4479 - val_accuracy: 0.8140 - 69ms/epoch - 4ms/step\n", "Epoch 310/400\n", "16/16 - 0s - loss: 0.4806 - accuracy: 0.8020 - val_loss: 0.4473 - val_accuracy: 0.8120 - 65ms/epoch - 4ms/step\n", "Epoch 311/400\n", "16/16 - 0s - loss: 0.4801 - accuracy: 0.8020 - val_loss: 0.4468 - val_accuracy: 0.8120 - 64ms/epoch - 4ms/step\n", "Epoch 312/400\n", "16/16 - 0s - loss: 0.4796 - accuracy: 0.8020 - val_loss: 0.4462 - val_accuracy: 0.8120 - 72ms/epoch - 4ms/step\n", "Epoch 313/400\n", "16/16 - 0s - loss: 0.4792 - accuracy: 0.8020 - val_loss: 0.4456 - val_accuracy: 0.8140 - 77ms/epoch - 5ms/step\n", "Epoch 314/400\n", "16/16 - 0s - loss: 0.4785 - accuracy: 0.8000 - val_loss: 0.4450 - val_accuracy: 0.8120 - 70ms/epoch - 4ms/step\n", "Epoch 315/400\n", "16/16 - 0s - loss: 0.4779 - accuracy: 0.7980 - val_loss: 0.4444 - val_accuracy: 0.8120 - 105ms/epoch - 7ms/step\n", "Epoch 316/400\n", "16/16 - 0s - loss: 0.4774 - accuracy: 0.8000 - val_loss: 0.4437 - val_accuracy: 0.8120 - 66ms/epoch - 4ms/step\n", "Epoch 317/400\n", "16/16 - 0s - loss: 0.4768 - accuracy: 0.7980 - val_loss: 0.4431 - val_accuracy: 0.8120 - 66ms/epoch - 4ms/step\n", "Epoch 318/400\n", "16/16 - 0s - loss: 0.4763 - accuracy: 0.7960 - val_loss: 0.4424 - val_accuracy: 0.8120 - 76ms/epoch - 5ms/step\n", "Epoch 319/400\n", "16/16 - 0s - loss: 0.4757 - accuracy: 0.7960 - val_loss: 0.4417 - val_accuracy: 0.8120 - 65ms/epoch - 4ms/step\n", "Epoch 320/400\n", "16/16 - 0s - loss: 0.4751 - accuracy: 0.7960 - val_loss: 0.4411 - val_accuracy: 0.8140 - 71ms/epoch - 4ms/step\n", "Epoch 321/400\n", "16/16 - 0s - loss: 0.4745 - accuracy: 0.7980 - val_loss: 0.4404 - val_accuracy: 0.8140 - 65ms/epoch - 4ms/step\n", "Epoch 322/400\n", "16/16 - 0s - loss: 0.4738 - accuracy: 0.8000 - val_loss: 0.4397 - val_accuracy: 0.8120 - 64ms/epoch - 4ms/step\n", "Epoch 323/400\n", "16/16 - 0s - loss: 0.4732 - accuracy: 0.8020 - val_loss: 0.4390 - val_accuracy: 0.8120 - 66ms/epoch - 4ms/step\n", "Epoch 324/400\n", "16/16 - 0s - loss: 0.4726 - accuracy: 0.8020 - val_loss: 0.4384 - val_accuracy: 0.8120 - 69ms/epoch - 4ms/step\n", "Epoch 325/400\n", "16/16 - 0s - loss: 0.4719 - accuracy: 0.8060 - val_loss: 0.4376 - val_accuracy: 0.8120 - 68ms/epoch - 4ms/step\n", "Epoch 326/400\n", "16/16 - 0s - loss: 0.4711 - accuracy: 0.8060 - val_loss: 0.4368 - val_accuracy: 0.8100 - 74ms/epoch - 5ms/step\n", "Epoch 327/400\n", "16/16 - 0s - loss: 0.4705 - accuracy: 0.8040 - val_loss: 0.4359 - val_accuracy: 0.8080 - 107ms/epoch - 7ms/step\n", "Epoch 328/400\n", "16/16 - 0s - loss: 0.4697 - accuracy: 0.8040 - val_loss: 0.4352 - val_accuracy: 0.8080 - 65ms/epoch - 4ms/step\n", "Epoch 329/400\n", "16/16 - 0s - loss: 0.4690 - accuracy: 0.8060 - val_loss: 0.4345 - val_accuracy: 0.8100 - 61ms/epoch - 4ms/step\n", "Epoch 330/400\n", "16/16 - 0s - loss: 0.4685 - accuracy: 0.8020 - val_loss: 0.4336 - val_accuracy: 0.8060 - 66ms/epoch - 4ms/step\n", "Epoch 331/400\n", "16/16 - 0s - loss: 0.4675 - accuracy: 0.8040 - val_loss: 0.4329 - val_accuracy: 0.8080 - 64ms/epoch - 4ms/step\n", "Epoch 332/400\n", "16/16 - 0s - loss: 0.4669 - accuracy: 0.8020 - val_loss: 0.4321 - val_accuracy: 0.8100 - 111ms/epoch - 7ms/step\n", "Epoch 333/400\n", "16/16 - 0s - loss: 0.4661 - accuracy: 0.8020 - val_loss: 0.4312 - val_accuracy: 0.8120 - 68ms/epoch - 4ms/step\n", "Epoch 334/400\n", "16/16 - 0s - loss: 0.4653 - accuracy: 0.8020 - val_loss: 0.4305 - val_accuracy: 0.8100 - 64ms/epoch - 4ms/step\n", "Epoch 335/400\n", "16/16 - 0s - loss: 0.4645 - accuracy: 0.8020 - val_loss: 0.4297 - val_accuracy: 0.8100 - 66ms/epoch - 4ms/step\n", "Epoch 336/400\n", "16/16 - 0s - loss: 0.4638 - accuracy: 0.8000 - val_loss: 0.4288 - val_accuracy: 0.8100 - 66ms/epoch - 4ms/step\n", "Epoch 337/400\n", "16/16 - 0s - loss: 0.4629 - accuracy: 0.8000 - val_loss: 0.4280 - val_accuracy: 0.8100 - 71ms/epoch - 4ms/step\n", "Epoch 338/400\n", "16/16 - 0s - loss: 0.4622 - accuracy: 0.8000 - val_loss: 0.4271 - val_accuracy: 0.8120 - 68ms/epoch - 4ms/step\n", "Epoch 339/400\n", "16/16 - 0s - loss: 0.4613 - accuracy: 0.8000 - val_loss: 0.4263 - val_accuracy: 0.8140 - 109ms/epoch - 7ms/step\n", "Epoch 340/400\n", "16/16 - 0s - loss: 0.4604 - accuracy: 0.8000 - val_loss: 0.4254 - val_accuracy: 0.8140 - 63ms/epoch - 4ms/step\n", "Epoch 341/400\n", "16/16 - 0s - loss: 0.4595 - accuracy: 0.8000 - val_loss: 0.4244 - val_accuracy: 0.8140 - 109ms/epoch - 7ms/step\n", "Epoch 342/400\n", "16/16 - 0s - loss: 0.4585 - accuracy: 0.8040 - val_loss: 0.4235 - val_accuracy: 0.8140 - 74ms/epoch - 5ms/step\n", "Epoch 343/400\n", "16/16 - 0s - loss: 0.4576 - accuracy: 0.8040 - val_loss: 0.4225 - val_accuracy: 0.8140 - 70ms/epoch - 4ms/step\n", "Epoch 344/400\n", "16/16 - 0s - loss: 0.4567 - accuracy: 0.8040 - val_loss: 0.4215 - val_accuracy: 0.8120 - 70ms/epoch - 4ms/step\n", "Epoch 345/400\n", "16/16 - 0s - loss: 0.4557 - accuracy: 0.8040 - val_loss: 0.4206 - val_accuracy: 0.8120 - 115ms/epoch - 7ms/step\n", "Epoch 346/400\n", "16/16 - 0s - loss: 0.4547 - accuracy: 0.8020 - val_loss: 0.4194 - val_accuracy: 0.8100 - 66ms/epoch - 4ms/step\n", "Epoch 347/400\n", "16/16 - 0s - loss: 0.4536 - accuracy: 0.8020 - val_loss: 0.4186 - val_accuracy: 0.8100 - 73ms/epoch - 5ms/step\n", "Epoch 348/400\n", "16/16 - 0s - loss: 0.4526 - accuracy: 0.8020 - val_loss: 0.4176 - val_accuracy: 0.8100 - 68ms/epoch - 4ms/step\n", "Epoch 349/400\n", "16/16 - 0s - loss: 0.4516 - accuracy: 0.8040 - val_loss: 0.4165 - val_accuracy: 0.8100 - 68ms/epoch - 4ms/step\n", "Epoch 350/400\n", "16/16 - 0s - loss: 0.4506 - accuracy: 0.8060 - val_loss: 0.4155 - val_accuracy: 0.8100 - 112ms/epoch - 7ms/step\n", "Epoch 351/400\n", "16/16 - 0s - loss: 0.4494 - accuracy: 0.8080 - val_loss: 0.4144 - val_accuracy: 0.8120 - 74ms/epoch - 5ms/step\n", "Epoch 352/400\n", "16/16 - 0s - loss: 0.4483 - accuracy: 0.8100 - val_loss: 0.4133 - val_accuracy: 0.8140 - 106ms/epoch - 7ms/step\n", "Epoch 353/400\n", "16/16 - 0s - loss: 0.4474 - accuracy: 0.8100 - val_loss: 0.4122 - val_accuracy: 0.8120 - 110ms/epoch - 7ms/step\n", "Epoch 354/400\n", "16/16 - 0s - loss: 0.4462 - accuracy: 0.8120 - val_loss: 0.4111 - val_accuracy: 0.8120 - 65ms/epoch - 4ms/step\n", "Epoch 355/400\n", "16/16 - 0s - loss: 0.4450 - accuracy: 0.8140 - val_loss: 0.4101 - val_accuracy: 0.8160 - 66ms/epoch - 4ms/step\n", "Epoch 356/400\n", "16/16 - 0s - loss: 0.4440 - accuracy: 0.8180 - val_loss: 0.4090 - val_accuracy: 0.8180 - 106ms/epoch - 7ms/step\n", "Epoch 357/400\n", "16/16 - 0s - loss: 0.4427 - accuracy: 0.8180 - val_loss: 0.4079 - val_accuracy: 0.8160 - 68ms/epoch - 4ms/step\n", "Epoch 358/400\n", "16/16 - 0s - loss: 0.4415 - accuracy: 0.8200 - val_loss: 0.4067 - val_accuracy: 0.8160 - 70ms/epoch - 4ms/step\n", "Epoch 359/400\n", "16/16 - 0s - loss: 0.4403 - accuracy: 0.8240 - val_loss: 0.4056 - val_accuracy: 0.8160 - 72ms/epoch - 4ms/step\n", "Epoch 360/400\n", "16/16 - 0s - loss: 0.4391 - accuracy: 0.8260 - val_loss: 0.4045 - val_accuracy: 0.8180 - 111ms/epoch - 7ms/step\n", "Epoch 361/400\n", "16/16 - 0s - loss: 0.4379 - accuracy: 0.8260 - val_loss: 0.4033 - val_accuracy: 0.8180 - 69ms/epoch - 4ms/step\n", "Epoch 362/400\n", "16/16 - 0s - loss: 0.4366 - accuracy: 0.8220 - val_loss: 0.4020 - val_accuracy: 0.8180 - 73ms/epoch - 5ms/step\n", "Epoch 363/400\n", "16/16 - 0s - loss: 0.4353 - accuracy: 0.8220 - val_loss: 0.4008 - val_accuracy: 0.8180 - 69ms/epoch - 4ms/step\n", "Epoch 364/400\n", "16/16 - 0s - loss: 0.4341 - accuracy: 0.8220 - val_loss: 0.3996 - val_accuracy: 0.8180 - 113ms/epoch - 7ms/step\n", "Epoch 365/400\n", "16/16 - 0s - loss: 0.4328 - accuracy: 0.8220 - val_loss: 0.3983 - val_accuracy: 0.8180 - 72ms/epoch - 5ms/step\n", "Epoch 366/400\n", "16/16 - 0s - loss: 0.4314 - accuracy: 0.8220 - val_loss: 0.3971 - val_accuracy: 0.8220 - 74ms/epoch - 5ms/step\n", "Epoch 367/400\n", "16/16 - 0s - loss: 0.4301 - accuracy: 0.8220 - val_loss: 0.3958 - val_accuracy: 0.8240 - 72ms/epoch - 4ms/step\n", "Epoch 368/400\n", "16/16 - 0s - loss: 0.4287 - accuracy: 0.8220 - val_loss: 0.3945 - val_accuracy: 0.8260 - 105ms/epoch - 7ms/step\n", "Epoch 369/400\n", "16/16 - 0s - loss: 0.4273 - accuracy: 0.8240 - val_loss: 0.3932 - val_accuracy: 0.8260 - 70ms/epoch - 4ms/step\n", "Epoch 370/400\n", "16/16 - 0s - loss: 0.4260 - accuracy: 0.8260 - val_loss: 0.3919 - val_accuracy: 0.8260 - 77ms/epoch - 5ms/step\n", "Epoch 371/400\n", "16/16 - 0s - loss: 0.4246 - accuracy: 0.8260 - val_loss: 0.3906 - val_accuracy: 0.8260 - 73ms/epoch - 5ms/step\n", "Epoch 372/400\n", "16/16 - 0s - loss: 0.4231 - accuracy: 0.8260 - val_loss: 0.3893 - val_accuracy: 0.8260 - 111ms/epoch - 7ms/step\n", "Epoch 373/400\n", "16/16 - 0s - loss: 0.4217 - accuracy: 0.8280 - val_loss: 0.3879 - val_accuracy: 0.8280 - 70ms/epoch - 4ms/step\n", "Epoch 374/400\n", "16/16 - 0s - loss: 0.4202 - accuracy: 0.8300 - val_loss: 0.3865 - val_accuracy: 0.8280 - 117ms/epoch - 7ms/step\n", "Epoch 375/400\n", "16/16 - 0s - loss: 0.4188 - accuracy: 0.8300 - val_loss: 0.3851 - val_accuracy: 0.8280 - 70ms/epoch - 4ms/step\n", "Epoch 376/400\n", "16/16 - 0s - loss: 0.4173 - accuracy: 0.8280 - val_loss: 0.3837 - val_accuracy: 0.8280 - 66ms/epoch - 4ms/step\n", "Epoch 377/400\n", "16/16 - 0s - loss: 0.4157 - accuracy: 0.8280 - val_loss: 0.3822 - val_accuracy: 0.8320 - 76ms/epoch - 5ms/step\n", "Epoch 378/400\n", "16/16 - 0s - loss: 0.4141 - accuracy: 0.8280 - val_loss: 0.3808 - val_accuracy: 0.8320 - 75ms/epoch - 5ms/step\n", "Epoch 379/400\n", "16/16 - 0s - loss: 0.4125 - accuracy: 0.8300 - val_loss: 0.3794 - val_accuracy: 0.8380 - 66ms/epoch - 4ms/step\n", "Epoch 380/400\n", "16/16 - 0s - loss: 0.4110 - accuracy: 0.8280 - val_loss: 0.3779 - val_accuracy: 0.8400 - 68ms/epoch - 4ms/step\n", "Epoch 381/400\n", "16/16 - 0s - loss: 0.4094 - accuracy: 0.8280 - val_loss: 0.3764 - val_accuracy: 0.8400 - 107ms/epoch - 7ms/step\n", "Epoch 382/400\n", "16/16 - 0s - loss: 0.4079 - accuracy: 0.8260 - val_loss: 0.3749 - val_accuracy: 0.8440 - 72ms/epoch - 4ms/step\n", "Epoch 383/400\n", "16/16 - 0s - loss: 0.4062 - accuracy: 0.8280 - val_loss: 0.3736 - val_accuracy: 0.8460 - 67ms/epoch - 4ms/step\n", "Epoch 384/400\n", "16/16 - 0s - loss: 0.4046 - accuracy: 0.8280 - val_loss: 0.3720 - val_accuracy: 0.8460 - 107ms/epoch - 7ms/step\n", "Epoch 385/400\n", "16/16 - 0s - loss: 0.4029 - accuracy: 0.8300 - val_loss: 0.3706 - val_accuracy: 0.8460 - 66ms/epoch - 4ms/step\n", "Epoch 386/400\n", "16/16 - 0s - loss: 0.4014 - accuracy: 0.8300 - val_loss: 0.3691 - val_accuracy: 0.8460 - 109ms/epoch - 7ms/step\n", "Epoch 387/400\n", "16/16 - 0s - loss: 0.3996 - accuracy: 0.8300 - val_loss: 0.3676 - val_accuracy: 0.8460 - 70ms/epoch - 4ms/step\n", "Epoch 388/400\n", "16/16 - 0s - loss: 0.3980 - accuracy: 0.8280 - val_loss: 0.3662 - val_accuracy: 0.8460 - 67ms/epoch - 4ms/step\n", "Epoch 389/400\n", "16/16 - 0s - loss: 0.3964 - accuracy: 0.8280 - val_loss: 0.3646 - val_accuracy: 0.8460 - 66ms/epoch - 4ms/step\n", "Epoch 390/400\n", "16/16 - 0s - loss: 0.3946 - accuracy: 0.8300 - val_loss: 0.3631 - val_accuracy: 0.8460 - 69ms/epoch - 4ms/step\n", "Epoch 391/400\n", "16/16 - 0s - loss: 0.3930 - accuracy: 0.8340 - val_loss: 0.3616 - val_accuracy: 0.8480 - 65ms/epoch - 4ms/step\n", "Epoch 392/400\n", "16/16 - 0s - loss: 0.3912 - accuracy: 0.8320 - val_loss: 0.3600 - val_accuracy: 0.8520 - 65ms/epoch - 4ms/step\n", "Epoch 393/400\n", "16/16 - 0s - loss: 0.3895 - accuracy: 0.8360 - val_loss: 0.3585 - val_accuracy: 0.8520 - 75ms/epoch - 5ms/step\n", "Epoch 394/400\n", "16/16 - 0s - loss: 0.3878 - accuracy: 0.8340 - val_loss: 0.3570 - val_accuracy: 0.8540 - 73ms/epoch - 5ms/step\n", "Epoch 395/400\n", "16/16 - 0s - loss: 0.3862 - accuracy: 0.8400 - val_loss: 0.3554 - val_accuracy: 0.8540 - 69ms/epoch - 4ms/step\n", "Epoch 396/400\n", "16/16 - 0s - loss: 0.3844 - accuracy: 0.8440 - val_loss: 0.3538 - val_accuracy: 0.8560 - 69ms/epoch - 4ms/step\n", "Epoch 397/400\n", "16/16 - 0s - loss: 0.3826 - accuracy: 0.8440 - val_loss: 0.3523 - val_accuracy: 0.8580 - 72ms/epoch - 4ms/step\n", "Epoch 398/400\n", "16/16 - 0s - loss: 0.3809 - accuracy: 0.8440 - val_loss: 0.3507 - val_accuracy: 0.8560 - 111ms/epoch - 7ms/step\n", "Epoch 399/400\n", "16/16 - 0s - loss: 0.3791 - accuracy: 0.8460 - val_loss: 0.3491 - val_accuracy: 0.8580 - 84ms/epoch - 5ms/step\n", "Epoch 400/400\n", "16/16 - 0s - loss: 0.3774 - accuracy: 0.8460 - val_loss: 0.3474 - val_accuracy: 0.8600 - 70ms/epoch - 4ms/step\n", "第1个弱分类器训练完毕\n", "Epoch 1/400\n", "16/16 - 1s - loss: 0.6834 - accuracy: 0.3760 - val_loss: 0.6751 - val_accuracy: 0.4420 - 801ms/epoch - 50ms/step\n", "Epoch 2/400\n", "16/16 - 0s - loss: 0.6795 - accuracy: 0.3960 - val_loss: 0.6706 - val_accuracy: 0.4600 - 111ms/epoch - 7ms/step\n", "Epoch 3/400\n", "16/16 - 0s - loss: 0.6759 - accuracy: 0.4040 - val_loss: 0.6665 - val_accuracy: 0.4720 - 73ms/epoch - 5ms/step\n", "Epoch 4/400\n", "16/16 - 0s - loss: 0.6724 - accuracy: 0.4200 - val_loss: 0.6627 - val_accuracy: 0.4900 - 70ms/epoch - 4ms/step\n", "Epoch 5/400\n", "16/16 - 0s - loss: 0.6692 - accuracy: 0.4300 - val_loss: 0.6587 - val_accuracy: 0.5140 - 70ms/epoch - 4ms/step\n", "Epoch 6/400\n", "16/16 - 0s - loss: 0.6658 - accuracy: 0.4520 - val_loss: 0.6551 - val_accuracy: 0.5260 - 67ms/epoch - 4ms/step\n", "Epoch 7/400\n", "16/16 - 0s - loss: 0.6627 - accuracy: 0.4720 - val_loss: 0.6518 - val_accuracy: 0.5260 - 81ms/epoch - 5ms/step\n", "Epoch 8/400\n", "16/16 - 0s - loss: 0.6596 - accuracy: 0.4740 - val_loss: 0.6483 - val_accuracy: 0.5260 - 109ms/epoch - 7ms/step\n", "Epoch 9/400\n", "16/16 - 0s - loss: 0.6565 - accuracy: 0.4740 - val_loss: 0.6448 - val_accuracy: 0.5260 - 67ms/epoch - 4ms/step\n", "Epoch 10/400\n", "16/16 - 0s - loss: 0.6534 - accuracy: 0.4740 - val_loss: 0.6415 - val_accuracy: 0.5260 - 70ms/epoch - 4ms/step\n", "Epoch 11/400\n", "16/16 - 0s - loss: 0.6504 - accuracy: 0.4740 - val_loss: 0.6380 - val_accuracy: 0.5260 - 68ms/epoch - 4ms/step\n", "Epoch 12/400\n", "16/16 - 0s - loss: 0.6474 - accuracy: 0.4840 - val_loss: 0.6349 - val_accuracy: 0.5360 - 117ms/epoch - 7ms/step\n", "Epoch 13/400\n", "16/16 - 0s - loss: 0.6445 - accuracy: 0.5080 - val_loss: 0.6317 - val_accuracy: 0.5600 - 78ms/epoch - 5ms/step\n", "Epoch 14/400\n", "16/16 - 0s - loss: 0.6416 - accuracy: 0.5460 - val_loss: 0.6283 - val_accuracy: 0.5900 - 111ms/epoch - 7ms/step\n", "Epoch 15/400\n", "16/16 - 0s - loss: 0.6386 - accuracy: 0.5740 - val_loss: 0.6252 - val_accuracy: 0.6120 - 66ms/epoch - 4ms/step\n", "Epoch 16/400\n", "16/16 - 0s - loss: 0.6357 - accuracy: 0.5900 - val_loss: 0.6222 - val_accuracy: 0.6280 - 77ms/epoch - 5ms/step\n", "Epoch 17/400\n", "16/16 - 0s - loss: 0.6329 - accuracy: 0.6000 - val_loss: 0.6191 - val_accuracy: 0.6300 - 111ms/epoch - 7ms/step\n", "Epoch 18/400\n", "16/16 - 0s - loss: 0.6300 - accuracy: 0.6100 - val_loss: 0.6160 - val_accuracy: 0.6400 - 111ms/epoch - 7ms/step\n", "Epoch 19/400\n", "16/16 - 0s - loss: 0.6272 - accuracy: 0.6160 - val_loss: 0.6129 - val_accuracy: 0.6400 - 114ms/epoch - 7ms/step\n", "Epoch 20/400\n", "16/16 - 0s - loss: 0.6243 - accuracy: 0.6120 - val_loss: 0.6099 - val_accuracy: 0.6300 - 72ms/epoch - 5ms/step\n", "Epoch 21/400\n", "16/16 - 0s - loss: 0.6215 - accuracy: 0.6180 - val_loss: 0.6069 - val_accuracy: 0.6280 - 108ms/epoch - 7ms/step\n", "Epoch 22/400\n", "16/16 - 0s - loss: 0.6187 - accuracy: 0.6180 - val_loss: 0.6039 - val_accuracy: 0.6280 - 68ms/epoch - 4ms/step\n", "Epoch 23/400\n", "16/16 - 0s - loss: 0.6159 - accuracy: 0.6160 - val_loss: 0.6008 - val_accuracy: 0.6260 - 109ms/epoch - 7ms/step\n", "Epoch 24/400\n", "16/16 - 0s - loss: 0.6131 - accuracy: 0.6040 - val_loss: 0.5979 - val_accuracy: 0.6220 - 109ms/epoch - 7ms/step\n", "Epoch 25/400\n", "16/16 - 0s - loss: 0.6103 - accuracy: 0.6000 - val_loss: 0.5950 - val_accuracy: 0.6220 - 64ms/epoch - 4ms/step\n", "Epoch 26/400\n", "16/16 - 0s - loss: 0.6076 - accuracy: 0.5960 - val_loss: 0.5920 - val_accuracy: 0.6140 - 71ms/epoch - 4ms/step\n", "Epoch 27/400\n", "16/16 - 0s - loss: 0.6048 - accuracy: 0.6020 - val_loss: 0.5892 - val_accuracy: 0.6040 - 65ms/epoch - 4ms/step\n", "Epoch 28/400\n", "16/16 - 0s - loss: 0.6021 - accuracy: 0.6020 - val_loss: 0.5863 - val_accuracy: 0.6040 - 67ms/epoch - 4ms/step\n", "Epoch 29/400\n", "16/16 - 0s - loss: 0.5993 - accuracy: 0.5980 - val_loss: 0.5834 - val_accuracy: 0.6120 - 68ms/epoch - 4ms/step\n", "Epoch 30/400\n", "16/16 - 0s - loss: 0.5967 - accuracy: 0.6000 - val_loss: 0.5806 - val_accuracy: 0.6140 - 78ms/epoch - 5ms/step\n", "Epoch 31/400\n", "16/16 - 0s - loss: 0.5939 - accuracy: 0.6020 - val_loss: 0.5778 - val_accuracy: 0.6280 - 112ms/epoch - 7ms/step\n", "Epoch 32/400\n", "16/16 - 0s - loss: 0.5912 - accuracy: 0.6060 - val_loss: 0.5751 - val_accuracy: 0.6360 - 111ms/epoch - 7ms/step\n", "Epoch 33/400\n", "16/16 - 0s - loss: 0.5885 - accuracy: 0.6100 - val_loss: 0.5721 - val_accuracy: 0.6380 - 71ms/epoch - 4ms/step\n", "Epoch 34/400\n", "16/16 - 0s - loss: 0.5858 - accuracy: 0.6160 - val_loss: 0.5693 - val_accuracy: 0.6500 - 112ms/epoch - 7ms/step\n", "Epoch 35/400\n", "16/16 - 0s - loss: 0.5831 - accuracy: 0.6200 - val_loss: 0.5665 - val_accuracy: 0.6520 - 114ms/epoch - 7ms/step\n", "Epoch 36/400\n", "16/16 - 0s - loss: 0.5804 - accuracy: 0.6240 - val_loss: 0.5636 - val_accuracy: 0.6500 - 71ms/epoch - 4ms/step\n", "Epoch 37/400\n", "16/16 - 0s - loss: 0.5777 - accuracy: 0.6260 - val_loss: 0.5607 - val_accuracy: 0.6540 - 109ms/epoch - 7ms/step\n", "Epoch 38/400\n", "16/16 - 0s - loss: 0.5750 - accuracy: 0.6360 - val_loss: 0.5579 - val_accuracy: 0.6520 - 66ms/epoch - 4ms/step\n", "Epoch 39/400\n", "16/16 - 0s - loss: 0.5724 - accuracy: 0.6360 - val_loss: 0.5551 - val_accuracy: 0.6540 - 69ms/epoch - 4ms/step\n", "Epoch 40/400\n", "16/16 - 0s - loss: 0.5696 - accuracy: 0.6440 - val_loss: 0.5523 - val_accuracy: 0.6600 - 106ms/epoch - 7ms/step\n", "Epoch 41/400\n", "16/16 - 0s - loss: 0.5670 - accuracy: 0.6540 - val_loss: 0.5495 - val_accuracy: 0.6620 - 72ms/epoch - 4ms/step\n", "Epoch 42/400\n", "16/16 - 0s - loss: 0.5643 - accuracy: 0.6560 - val_loss: 0.5467 - val_accuracy: 0.6680 - 105ms/epoch - 7ms/step\n", "Epoch 43/400\n", "16/16 - 0s - loss: 0.5616 - accuracy: 0.6640 - val_loss: 0.5438 - val_accuracy: 0.6700 - 76ms/epoch - 5ms/step\n", "Epoch 44/400\n", "16/16 - 0s - loss: 0.5589 - accuracy: 0.6640 - val_loss: 0.5410 - val_accuracy: 0.6700 - 71ms/epoch - 4ms/step\n", "Epoch 45/400\n", "16/16 - 0s - loss: 0.5562 - accuracy: 0.6680 - val_loss: 0.5383 - val_accuracy: 0.6740 - 71ms/epoch - 4ms/step\n", "Epoch 46/400\n", "16/16 - 0s - loss: 0.5535 - accuracy: 0.6680 - val_loss: 0.5355 - val_accuracy: 0.6800 - 122ms/epoch - 8ms/step\n", "Epoch 47/400\n", "16/16 - 0s - loss: 0.5508 - accuracy: 0.6740 - val_loss: 0.5326 - val_accuracy: 0.6820 - 70ms/epoch - 4ms/step\n", "Epoch 48/400\n", "16/16 - 0s - loss: 0.5481 - accuracy: 0.6800 - val_loss: 0.5299 - val_accuracy: 0.6840 - 64ms/epoch - 4ms/step\n", "Epoch 49/400\n", "16/16 - 0s - loss: 0.5454 - accuracy: 0.6860 - val_loss: 0.5271 - val_accuracy: 0.6860 - 113ms/epoch - 7ms/step\n", "Epoch 50/400\n", "16/16 - 0s - loss: 0.5427 - accuracy: 0.6900 - val_loss: 0.5244 - val_accuracy: 0.6960 - 70ms/epoch - 4ms/step\n", "Epoch 51/400\n", "16/16 - 0s - loss: 0.5399 - accuracy: 0.6960 - val_loss: 0.5215 - val_accuracy: 0.6960 - 73ms/epoch - 5ms/step\n", "Epoch 52/400\n", "16/16 - 0s - loss: 0.5372 - accuracy: 0.6960 - val_loss: 0.5186 - val_accuracy: 0.7020 - 67ms/epoch - 4ms/step\n", "Epoch 53/400\n", "16/16 - 0s - loss: 0.5345 - accuracy: 0.6980 - val_loss: 0.5158 - val_accuracy: 0.7100 - 113ms/epoch - 7ms/step\n", "Epoch 54/400\n", "16/16 - 0s - loss: 0.5318 - accuracy: 0.7080 - val_loss: 0.5130 - val_accuracy: 0.7180 - 69ms/epoch - 4ms/step\n", "Epoch 55/400\n", "16/16 - 0s - loss: 0.5290 - accuracy: 0.7140 - val_loss: 0.5102 - val_accuracy: 0.7260 - 75ms/epoch - 5ms/step\n", "Epoch 56/400\n", "16/16 - 0s - loss: 0.5263 - accuracy: 0.7140 - val_loss: 0.5074 - val_accuracy: 0.7380 - 71ms/epoch - 4ms/step\n", "Epoch 57/400\n", "16/16 - 0s - loss: 0.5236 - accuracy: 0.7160 - val_loss: 0.5046 - val_accuracy: 0.7420 - 81ms/epoch - 5ms/step\n", "Epoch 58/400\n", "16/16 - 0s - loss: 0.5208 - accuracy: 0.7200 - val_loss: 0.5017 - val_accuracy: 0.7400 - 70ms/epoch - 4ms/step\n", "Epoch 59/400\n", "16/16 - 0s - loss: 0.5181 - accuracy: 0.7260 - val_loss: 0.4988 - val_accuracy: 0.7420 - 80ms/epoch - 5ms/step\n", "Epoch 60/400\n", "16/16 - 0s - loss: 0.5154 - accuracy: 0.7320 - val_loss: 0.4961 - val_accuracy: 0.7440 - 72ms/epoch - 4ms/step\n", "Epoch 61/400\n", "16/16 - 0s - loss: 0.5126 - accuracy: 0.7400 - val_loss: 0.4933 - val_accuracy: 0.7520 - 111ms/epoch - 7ms/step\n", "Epoch 62/400\n", "16/16 - 0s - loss: 0.5098 - accuracy: 0.7440 - val_loss: 0.4904 - val_accuracy: 0.7600 - 68ms/epoch - 4ms/step\n", "Epoch 63/400\n", "16/16 - 0s - loss: 0.5070 - accuracy: 0.7440 - val_loss: 0.4875 - val_accuracy: 0.7600 - 110ms/epoch - 7ms/step\n", "Epoch 64/400\n", "16/16 - 0s - loss: 0.5042 - accuracy: 0.7500 - val_loss: 0.4847 - val_accuracy: 0.7640 - 68ms/epoch - 4ms/step\n", "Epoch 65/400\n", "16/16 - 0s - loss: 0.5014 - accuracy: 0.7540 - val_loss: 0.4818 - val_accuracy: 0.7660 - 112ms/epoch - 7ms/step\n", "Epoch 66/400\n", "16/16 - 0s - loss: 0.4986 - accuracy: 0.7640 - val_loss: 0.4789 - val_accuracy: 0.7720 - 110ms/epoch - 7ms/step\n", "Epoch 67/400\n", "16/16 - 0s - loss: 0.4958 - accuracy: 0.7720 - val_loss: 0.4761 - val_accuracy: 0.7780 - 115ms/epoch - 7ms/step\n", "Epoch 68/400\n", "16/16 - 0s - loss: 0.4929 - accuracy: 0.7760 - val_loss: 0.4731 - val_accuracy: 0.7780 - 117ms/epoch - 7ms/step\n", "Epoch 69/400\n", "16/16 - 0s - loss: 0.4900 - accuracy: 0.7780 - val_loss: 0.4702 - val_accuracy: 0.7800 - 68ms/epoch - 4ms/step\n", "Epoch 70/400\n", "16/16 - 0s - loss: 0.4871 - accuracy: 0.7860 - val_loss: 0.4673 - val_accuracy: 0.7840 - 119ms/epoch - 7ms/step\n", "Epoch 71/400\n", "16/16 - 0s - loss: 0.4842 - accuracy: 0.7900 - val_loss: 0.4643 - val_accuracy: 0.7920 - 72ms/epoch - 4ms/step\n", "Epoch 72/400\n", "16/16 - 0s - loss: 0.4812 - accuracy: 0.8020 - val_loss: 0.4614 - val_accuracy: 0.7980 - 78ms/epoch - 5ms/step\n", "Epoch 73/400\n", "16/16 - 0s - loss: 0.4783 - accuracy: 0.8080 - val_loss: 0.4585 - val_accuracy: 0.8080 - 79ms/epoch - 5ms/step\n", "Epoch 74/400\n", "16/16 - 0s - loss: 0.4753 - accuracy: 0.8100 - val_loss: 0.4556 - val_accuracy: 0.8140 - 108ms/epoch - 7ms/step\n", "Epoch 75/400\n", "16/16 - 0s - loss: 0.4723 - accuracy: 0.8180 - val_loss: 0.4526 - val_accuracy: 0.8220 - 67ms/epoch - 4ms/step\n", "Epoch 76/400\n", "16/16 - 0s - loss: 0.4694 - accuracy: 0.8180 - val_loss: 0.4496 - val_accuracy: 0.8240 - 78ms/epoch - 5ms/step\n", "Epoch 77/400\n", "16/16 - 0s - loss: 0.4664 - accuracy: 0.8240 - val_loss: 0.4466 - val_accuracy: 0.8280 - 67ms/epoch - 4ms/step\n", "Epoch 78/400\n", "16/16 - 0s - loss: 0.4634 - accuracy: 0.8260 - val_loss: 0.4437 - val_accuracy: 0.8380 - 67ms/epoch - 4ms/step\n", "Epoch 79/400\n", "16/16 - 0s - loss: 0.4603 - accuracy: 0.8340 - val_loss: 0.4407 - val_accuracy: 0.8400 - 84ms/epoch - 5ms/step\n", "Epoch 80/400\n", "16/16 - 0s - loss: 0.4573 - accuracy: 0.8400 - val_loss: 0.4376 - val_accuracy: 0.8420 - 81ms/epoch - 5ms/step\n", "Epoch 81/400\n", "16/16 - 0s - loss: 0.4542 - accuracy: 0.8460 - val_loss: 0.4348 - val_accuracy: 0.8460 - 66ms/epoch - 4ms/step\n", "Epoch 82/400\n", "16/16 - 0s - loss: 0.4512 - accuracy: 0.8500 - val_loss: 0.4318 - val_accuracy: 0.8500 - 70ms/epoch - 4ms/step\n", "Epoch 83/400\n", "16/16 - 0s - loss: 0.4481 - accuracy: 0.8560 - val_loss: 0.4287 - val_accuracy: 0.8560 - 68ms/epoch - 4ms/step\n", "Epoch 84/400\n", "16/16 - 0s - loss: 0.4450 - accuracy: 0.8600 - val_loss: 0.4257 - val_accuracy: 0.8600 - 71ms/epoch - 4ms/step\n", "Epoch 85/400\n", "16/16 - 0s - loss: 0.4419 - accuracy: 0.8640 - val_loss: 0.4226 - val_accuracy: 0.8620 - 70ms/epoch - 4ms/step\n", "Epoch 86/400\n", "16/16 - 0s - loss: 0.4388 - accuracy: 0.8700 - val_loss: 0.4196 - val_accuracy: 0.8660 - 73ms/epoch - 5ms/step\n", "Epoch 87/400\n", "16/16 - 0s - loss: 0.4356 - accuracy: 0.8800 - val_loss: 0.4165 - val_accuracy: 0.8740 - 72ms/epoch - 4ms/step\n", "Epoch 88/400\n", "16/16 - 0s - loss: 0.4324 - accuracy: 0.8820 - val_loss: 0.4134 - val_accuracy: 0.8820 - 108ms/epoch - 7ms/step\n", "Epoch 89/400\n", "16/16 - 0s - loss: 0.4292 - accuracy: 0.8820 - val_loss: 0.4103 - val_accuracy: 0.8900 - 71ms/epoch - 4ms/step\n", "Epoch 90/400\n", "16/16 - 0s - loss: 0.4260 - accuracy: 0.8840 - val_loss: 0.4073 - val_accuracy: 0.8920 - 72ms/epoch - 5ms/step\n", "Epoch 91/400\n", "16/16 - 0s - loss: 0.4228 - accuracy: 0.8900 - val_loss: 0.4042 - val_accuracy: 0.8940 - 111ms/epoch - 7ms/step\n", "Epoch 92/400\n", "16/16 - 0s - loss: 0.4195 - accuracy: 0.8960 - val_loss: 0.4011 - val_accuracy: 0.9000 - 77ms/epoch - 5ms/step\n", "Epoch 93/400\n", "16/16 - 0s - loss: 0.4163 - accuracy: 0.8960 - val_loss: 0.3981 - val_accuracy: 0.9080 - 70ms/epoch - 4ms/step\n", "Epoch 94/400\n", "16/16 - 0s - loss: 0.4130 - accuracy: 0.9020 - val_loss: 0.3949 - val_accuracy: 0.9120 - 67ms/epoch - 4ms/step\n", "Epoch 95/400\n", "16/16 - 0s - loss: 0.4098 - accuracy: 0.9040 - val_loss: 0.3918 - val_accuracy: 0.9140 - 66ms/epoch - 4ms/step\n", "Epoch 96/400\n", "16/16 - 0s - loss: 0.4065 - accuracy: 0.9060 - val_loss: 0.3887 - val_accuracy: 0.9140 - 111ms/epoch - 7ms/step\n", "Epoch 97/400\n", "16/16 - 0s - loss: 0.4032 - accuracy: 0.9080 - val_loss: 0.3856 - val_accuracy: 0.9180 - 72ms/epoch - 4ms/step\n", "Epoch 98/400\n", "16/16 - 0s - loss: 0.4000 - accuracy: 0.9100 - val_loss: 0.3825 - val_accuracy: 0.9200 - 67ms/epoch - 4ms/step\n", "Epoch 99/400\n", "16/16 - 0s - loss: 0.3967 - accuracy: 0.9100 - val_loss: 0.3795 - val_accuracy: 0.9260 - 112ms/epoch - 7ms/step\n", "Epoch 100/400\n", "16/16 - 0s - loss: 0.3934 - accuracy: 0.9100 - val_loss: 0.3764 - val_accuracy: 0.9280 - 111ms/epoch - 7ms/step\n", "Epoch 101/400\n", "16/16 - 0s - loss: 0.3901 - accuracy: 0.9100 - val_loss: 0.3733 - val_accuracy: 0.9320 - 117ms/epoch - 7ms/step\n", "Epoch 102/400\n", "16/16 - 0s - loss: 0.3869 - accuracy: 0.9100 - val_loss: 0.3702 - val_accuracy: 0.9340 - 67ms/epoch - 4ms/step\n", "Epoch 103/400\n", "16/16 - 0s - loss: 0.3836 - accuracy: 0.9140 - val_loss: 0.3671 - val_accuracy: 0.9340 - 112ms/epoch - 7ms/step\n", "Epoch 104/400\n", "16/16 - 0s - loss: 0.3803 - accuracy: 0.9160 - val_loss: 0.3642 - val_accuracy: 0.9380 - 75ms/epoch - 5ms/step\n", "Epoch 105/400\n", "16/16 - 0s - loss: 0.3770 - accuracy: 0.9200 - val_loss: 0.3610 - val_accuracy: 0.9400 - 67ms/epoch - 4ms/step\n", "Epoch 106/400\n", "16/16 - 0s - loss: 0.3737 - accuracy: 0.9200 - val_loss: 0.3580 - val_accuracy: 0.9420 - 108ms/epoch - 7ms/step\n", "Epoch 107/400\n", "16/16 - 0s - loss: 0.3703 - accuracy: 0.9220 - val_loss: 0.3550 - val_accuracy: 0.9420 - 69ms/epoch - 4ms/step\n", "Epoch 108/400\n", "16/16 - 0s - loss: 0.3670 - accuracy: 0.9280 - val_loss: 0.3519 - val_accuracy: 0.9420 - 113ms/epoch - 7ms/step\n", "Epoch 109/400\n", "16/16 - 0s - loss: 0.3638 - accuracy: 0.9320 - val_loss: 0.3489 - val_accuracy: 0.9440 - 75ms/epoch - 5ms/step\n", "Epoch 110/400\n", "16/16 - 0s - loss: 0.3605 - accuracy: 0.9320 - val_loss: 0.3459 - val_accuracy: 0.9460 - 67ms/epoch - 4ms/step\n", "Epoch 111/400\n", "16/16 - 0s - loss: 0.3573 - accuracy: 0.9360 - val_loss: 0.3428 - val_accuracy: 0.9480 - 75ms/epoch - 5ms/step\n", "Epoch 112/400\n", "16/16 - 0s - loss: 0.3540 - accuracy: 0.9360 - val_loss: 0.3399 - val_accuracy: 0.9500 - 67ms/epoch - 4ms/step\n", "Epoch 113/400\n", "16/16 - 0s - loss: 0.3508 - accuracy: 0.9360 - val_loss: 0.3370 - val_accuracy: 0.9520 - 64ms/epoch - 4ms/step\n", "Epoch 114/400\n", "16/16 - 0s - loss: 0.3476 - accuracy: 0.9380 - val_loss: 0.3340 - val_accuracy: 0.9520 - 69ms/epoch - 4ms/step\n", "Epoch 115/400\n", "16/16 - 0s - loss: 0.3444 - accuracy: 0.9420 - val_loss: 0.3311 - val_accuracy: 0.9560 - 112ms/epoch - 7ms/step\n", "Epoch 116/400\n", "16/16 - 0s - loss: 0.3412 - accuracy: 0.9400 - val_loss: 0.3281 - val_accuracy: 0.9560 - 72ms/epoch - 5ms/step\n", "Epoch 117/400\n", "16/16 - 0s - loss: 0.3380 - accuracy: 0.9460 - val_loss: 0.3252 - val_accuracy: 0.9580 - 81ms/epoch - 5ms/step\n", "Epoch 118/400\n", "16/16 - 0s - loss: 0.3350 - accuracy: 0.9460 - val_loss: 0.3222 - val_accuracy: 0.9600 - 67ms/epoch - 4ms/step\n", "Epoch 119/400\n", "16/16 - 0s - loss: 0.3318 - accuracy: 0.9500 - val_loss: 0.3193 - val_accuracy: 0.9620 - 68ms/epoch - 4ms/step\n", "Epoch 120/400\n", "16/16 - 0s - loss: 0.3286 - accuracy: 0.9520 - val_loss: 0.3164 - val_accuracy: 0.9620 - 68ms/epoch - 4ms/step\n", "Epoch 121/400\n", "16/16 - 0s - loss: 0.3255 - accuracy: 0.9520 - val_loss: 0.3135 - val_accuracy: 0.9660 - 71ms/epoch - 4ms/step\n", "Epoch 122/400\n", "16/16 - 0s - loss: 0.3225 - accuracy: 0.9520 - val_loss: 0.3106 - val_accuracy: 0.9700 - 68ms/epoch - 4ms/step\n", "Epoch 123/400\n", "16/16 - 0s - loss: 0.3194 - accuracy: 0.9560 - val_loss: 0.3078 - val_accuracy: 0.9740 - 70ms/epoch - 4ms/step\n", "Epoch 124/400\n", "16/16 - 0s - loss: 0.3164 - accuracy: 0.9600 - val_loss: 0.3050 - val_accuracy: 0.9740 - 67ms/epoch - 4ms/step\n", "Epoch 125/400\n", "16/16 - 0s - loss: 0.3133 - accuracy: 0.9620 - val_loss: 0.3022 - val_accuracy: 0.9740 - 66ms/epoch - 4ms/step\n", "Epoch 126/400\n", "16/16 - 0s - loss: 0.3103 - accuracy: 0.9660 - val_loss: 0.2993 - val_accuracy: 0.9760 - 67ms/epoch - 4ms/step\n", "Epoch 127/400\n", "16/16 - 0s - loss: 0.3072 - accuracy: 0.9660 - val_loss: 0.2966 - val_accuracy: 0.9760 - 111ms/epoch - 7ms/step\n", "Epoch 128/400\n", "16/16 - 0s - loss: 0.3043 - accuracy: 0.9680 - val_loss: 0.2938 - val_accuracy: 0.9760 - 79ms/epoch - 5ms/step\n", "Epoch 129/400\n", "16/16 - 0s - loss: 0.3013 - accuracy: 0.9700 - val_loss: 0.2910 - val_accuracy: 0.9760 - 73ms/epoch - 5ms/step\n", "Epoch 130/400\n", "16/16 - 0s - loss: 0.2983 - accuracy: 0.9720 - val_loss: 0.2883 - val_accuracy: 0.9780 - 75ms/epoch - 5ms/step\n", "Epoch 131/400\n", "16/16 - 0s - loss: 0.2954 - accuracy: 0.9720 - val_loss: 0.2856 - val_accuracy: 0.9780 - 69ms/epoch - 4ms/step\n", "Epoch 132/400\n", "16/16 - 0s - loss: 0.2925 - accuracy: 0.9700 - val_loss: 0.2829 - val_accuracy: 0.9800 - 112ms/epoch - 7ms/step\n", "Epoch 133/400\n", "16/16 - 0s - loss: 0.2896 - accuracy: 0.9700 - val_loss: 0.2802 - val_accuracy: 0.9800 - 110ms/epoch - 7ms/step\n", "Epoch 134/400\n", "16/16 - 0s - loss: 0.2867 - accuracy: 0.9700 - val_loss: 0.2775 - val_accuracy: 0.9820 - 68ms/epoch - 4ms/step\n", "Epoch 135/400\n", "16/16 - 0s - loss: 0.2838 - accuracy: 0.9700 - val_loss: 0.2749 - val_accuracy: 0.9840 - 108ms/epoch - 7ms/step\n", "Epoch 136/400\n", "16/16 - 0s - loss: 0.2809 - accuracy: 0.9700 - val_loss: 0.2723 - val_accuracy: 0.9860 - 68ms/epoch - 4ms/step\n", "Epoch 137/400\n", "16/16 - 0s - loss: 0.2781 - accuracy: 0.9740 - val_loss: 0.2696 - val_accuracy: 0.9880 - 72ms/epoch - 5ms/step\n", "Epoch 138/400\n", "16/16 - 0s - loss: 0.2753 - accuracy: 0.9760 - val_loss: 0.2670 - val_accuracy: 0.9880 - 71ms/epoch - 4ms/step\n", "Epoch 139/400\n", "16/16 - 0s - loss: 0.2724 - accuracy: 0.9780 - val_loss: 0.2643 - val_accuracy: 0.9880 - 106ms/epoch - 7ms/step\n", "Epoch 140/400\n", "16/16 - 0s - loss: 0.2696 - accuracy: 0.9780 - val_loss: 0.2617 - val_accuracy: 0.9880 - 72ms/epoch - 5ms/step\n", "Epoch 141/400\n", "16/16 - 0s - loss: 0.2668 - accuracy: 0.9780 - val_loss: 0.2591 - val_accuracy: 0.9900 - 113ms/epoch - 7ms/step\n", "Epoch 142/400\n", "16/16 - 0s - loss: 0.2640 - accuracy: 0.9780 - val_loss: 0.2565 - val_accuracy: 0.9900 - 112ms/epoch - 7ms/step\n", "Epoch 143/400\n", "16/16 - 0s - loss: 0.2611 - accuracy: 0.9800 - val_loss: 0.2540 - val_accuracy: 0.9900 - 72ms/epoch - 4ms/step\n", "Epoch 144/400\n", "16/16 - 0s - loss: 0.2584 - accuracy: 0.9800 - val_loss: 0.2515 - val_accuracy: 0.9920 - 71ms/epoch - 4ms/step\n", "Epoch 145/400\n", "16/16 - 0s - loss: 0.2557 - accuracy: 0.9820 - val_loss: 0.2489 - val_accuracy: 0.9920 - 75ms/epoch - 5ms/step\n", "Epoch 146/400\n", "16/16 - 0s - loss: 0.2530 - accuracy: 0.9820 - val_loss: 0.2465 - val_accuracy: 0.9920 - 76ms/epoch - 5ms/step\n", "Epoch 147/400\n", "16/16 - 0s - loss: 0.2503 - accuracy: 0.9840 - val_loss: 0.2440 - val_accuracy: 0.9940 - 110ms/epoch - 7ms/step\n", "Epoch 148/400\n", "16/16 - 0s - loss: 0.2476 - accuracy: 0.9860 - val_loss: 0.2415 - val_accuracy: 0.9940 - 67ms/epoch - 4ms/step\n", "Epoch 149/400\n", "16/16 - 0s - loss: 0.2449 - accuracy: 0.9860 - val_loss: 0.2390 - val_accuracy: 0.9940 - 110ms/epoch - 7ms/step\n", "Epoch 150/400\n", "16/16 - 0s - loss: 0.2423 - accuracy: 0.9880 - val_loss: 0.2365 - val_accuracy: 0.9940 - 77ms/epoch - 5ms/step\n", "Epoch 151/400\n", "16/16 - 0s - loss: 0.2396 - accuracy: 0.9880 - val_loss: 0.2341 - val_accuracy: 0.9960 - 67ms/epoch - 4ms/step\n", "Epoch 152/400\n", "16/16 - 0s - loss: 0.2370 - accuracy: 0.9880 - val_loss: 0.2317 - val_accuracy: 0.9980 - 72ms/epoch - 4ms/step\n", "Epoch 153/400\n", "16/16 - 0s - loss: 0.2343 - accuracy: 0.9880 - val_loss: 0.2293 - val_accuracy: 0.9980 - 71ms/epoch - 4ms/step\n", "Epoch 154/400\n", "16/16 - 0s - loss: 0.2318 - accuracy: 0.9880 - val_loss: 0.2269 - val_accuracy: 0.9980 - 110ms/epoch - 7ms/step\n", "Epoch 155/400\n", "16/16 - 0s - loss: 0.2292 - accuracy: 0.9900 - val_loss: 0.2246 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 156/400\n", "16/16 - 0s - loss: 0.2267 - accuracy: 0.9920 - val_loss: 0.2222 - val_accuracy: 0.9980 - 109ms/epoch - 7ms/step\n", "Epoch 157/400\n", "16/16 - 0s - loss: 0.2242 - accuracy: 0.9920 - val_loss: 0.2199 - val_accuracy: 1.0000 - 108ms/epoch - 7ms/step\n", "Epoch 158/400\n", "16/16 - 0s - loss: 0.2217 - accuracy: 0.9920 - val_loss: 0.2177 - val_accuracy: 1.0000 - 67ms/epoch - 4ms/step\n", "Epoch 159/400\n", "16/16 - 0s - loss: 0.2192 - accuracy: 0.9920 - val_loss: 0.2154 - val_accuracy: 1.0000 - 109ms/epoch - 7ms/step\n", "Epoch 160/400\n", "16/16 - 0s - loss: 0.2168 - accuracy: 0.9920 - val_loss: 0.2132 - val_accuracy: 1.0000 - 69ms/epoch - 4ms/step\n", "Epoch 161/400\n", "16/16 - 0s - loss: 0.2145 - accuracy: 0.9920 - val_loss: 0.2110 - val_accuracy: 1.0000 - 69ms/epoch - 4ms/step\n", "Epoch 162/400\n", "16/16 - 0s - loss: 0.2121 - accuracy: 0.9940 - val_loss: 0.2088 - val_accuracy: 1.0000 - 123ms/epoch - 8ms/step\n", "Epoch 163/400\n", "16/16 - 0s - loss: 0.2097 - accuracy: 0.9940 - val_loss: 0.2066 - val_accuracy: 1.0000 - 67ms/epoch - 4ms/step\n", "Epoch 164/400\n", "16/16 - 0s - loss: 0.2074 - accuracy: 0.9940 - val_loss: 0.2045 - val_accuracy: 0.9980 - 67ms/epoch - 4ms/step\n", "Epoch 165/400\n", "16/16 - 0s - loss: 0.2052 - accuracy: 0.9940 - val_loss: 0.2024 - val_accuracy: 0.9980 - 73ms/epoch - 5ms/step\n", "Epoch 166/400\n", "16/16 - 0s - loss: 0.2029 - accuracy: 0.9940 - val_loss: 0.2004 - val_accuracy: 0.9980 - 66ms/epoch - 4ms/step\n", "Epoch 167/400\n", "16/16 - 0s - loss: 0.2007 - accuracy: 0.9960 - val_loss: 0.1983 - val_accuracy: 0.9980 - 72ms/epoch - 4ms/step\n", "Epoch 168/400\n", "16/16 - 0s - loss: 0.1986 - accuracy: 0.9960 - val_loss: 0.1963 - val_accuracy: 0.9980 - 68ms/epoch - 4ms/step\n", "Epoch 169/400\n", "16/16 - 0s - loss: 0.1963 - accuracy: 0.9960 - val_loss: 0.1943 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 170/400\n", "16/16 - 0s - loss: 0.1942 - accuracy: 0.9960 - val_loss: 0.1923 - val_accuracy: 0.9980 - 72ms/epoch - 5ms/step\n", "Epoch 171/400\n", "16/16 - 0s - loss: 0.1921 - accuracy: 0.9960 - val_loss: 0.1903 - val_accuracy: 0.9980 - 108ms/epoch - 7ms/step\n", "Epoch 172/400\n", "16/16 - 0s - loss: 0.1900 - accuracy: 0.9960 - val_loss: 0.1883 - val_accuracy: 0.9980 - 66ms/epoch - 4ms/step\n", "Epoch 173/400\n", "16/16 - 0s - loss: 0.1879 - accuracy: 0.9960 - val_loss: 0.1864 - val_accuracy: 0.9980 - 67ms/epoch - 4ms/step\n", "Epoch 174/400\n", "16/16 - 0s - loss: 0.1859 - accuracy: 0.9960 - val_loss: 0.1845 - val_accuracy: 0.9980 - 85ms/epoch - 5ms/step\n", "Epoch 175/400\n", "16/16 - 0s - loss: 0.1838 - accuracy: 0.9960 - val_loss: 0.1826 - val_accuracy: 0.9980 - 111ms/epoch - 7ms/step\n", "Epoch 176/400\n", "16/16 - 0s - loss: 0.1819 - accuracy: 0.9960 - val_loss: 0.1808 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 177/400\n", "16/16 - 0s - loss: 0.1799 - accuracy: 0.9960 - val_loss: 0.1790 - val_accuracy: 0.9980 - 113ms/epoch - 7ms/step\n", "Epoch 178/400\n", "16/16 - 0s - loss: 0.1780 - accuracy: 0.9980 - val_loss: 0.1773 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 179/400\n", "16/16 - 0s - loss: 0.1761 - accuracy: 0.9980 - val_loss: 0.1755 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 180/400\n", "16/16 - 0s - loss: 0.1743 - accuracy: 0.9980 - val_loss: 0.1738 - val_accuracy: 0.9980 - 73ms/epoch - 5ms/step\n", "Epoch 181/400\n", "16/16 - 0s - loss: 0.1724 - accuracy: 0.9980 - val_loss: 0.1721 - val_accuracy: 0.9980 - 66ms/epoch - 4ms/step\n", "Epoch 182/400\n", "16/16 - 0s - loss: 0.1706 - accuracy: 0.9980 - val_loss: 0.1705 - val_accuracy: 0.9980 - 67ms/epoch - 4ms/step\n", "Epoch 183/400\n", "16/16 - 0s - loss: 0.1689 - accuracy: 0.9980 - val_loss: 0.1688 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 184/400\n", "16/16 - 0s - loss: 0.1671 - accuracy: 0.9980 - val_loss: 0.1672 - val_accuracy: 0.9980 - 68ms/epoch - 4ms/step\n", "Epoch 185/400\n", "16/16 - 0s - loss: 0.1653 - accuracy: 0.9980 - val_loss: 0.1656 - val_accuracy: 0.9980 - 78ms/epoch - 5ms/step\n", "Epoch 186/400\n", "16/16 - 0s - loss: 0.1636 - accuracy: 0.9980 - val_loss: 0.1640 - val_accuracy: 0.9980 - 77ms/epoch - 5ms/step\n", "Epoch 187/400\n", "16/16 - 0s - loss: 0.1619 - accuracy: 0.9980 - val_loss: 0.1624 - val_accuracy: 0.9980 - 68ms/epoch - 4ms/step\n", "Epoch 188/400\n", "16/16 - 0s - loss: 0.1601 - accuracy: 0.9980 - val_loss: 0.1608 - val_accuracy: 0.9980 - 67ms/epoch - 4ms/step\n", "Epoch 189/400\n", "16/16 - 0s - loss: 0.1584 - accuracy: 0.9980 - val_loss: 0.1593 - val_accuracy: 0.9980 - 73ms/epoch - 5ms/step\n", "Epoch 190/400\n", "16/16 - 0s - loss: 0.1568 - accuracy: 0.9980 - val_loss: 0.1578 - val_accuracy: 0.9980 - 76ms/epoch - 5ms/step\n", "Epoch 191/400\n", "16/16 - 0s - loss: 0.1552 - accuracy: 0.9980 - val_loss: 0.1563 - val_accuracy: 0.9980 - 114ms/epoch - 7ms/step\n", "Epoch 192/400\n", "16/16 - 0s - loss: 0.1535 - accuracy: 0.9980 - val_loss: 0.1548 - val_accuracy: 0.9980 - 110ms/epoch - 7ms/step\n", "Epoch 193/400\n", "16/16 - 0s - loss: 0.1519 - accuracy: 0.9980 - val_loss: 0.1533 - val_accuracy: 0.9980 - 110ms/epoch - 7ms/step\n", "Epoch 194/400\n", "16/16 - 0s - loss: 0.1503 - accuracy: 0.9980 - val_loss: 0.1518 - val_accuracy: 0.9980 - 67ms/epoch - 4ms/step\n", "Epoch 195/400\n", "16/16 - 0s - loss: 0.1487 - accuracy: 0.9980 - val_loss: 0.1504 - val_accuracy: 0.9980 - 115ms/epoch - 7ms/step\n", "Epoch 196/400\n", "16/16 - 0s - loss: 0.1471 - accuracy: 0.9980 - val_loss: 0.1489 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 197/400\n", "16/16 - 0s - loss: 0.1456 - accuracy: 0.9980 - val_loss: 0.1475 - val_accuracy: 0.9980 - 68ms/epoch - 4ms/step\n", "Epoch 198/400\n", "16/16 - 0s - loss: 0.1441 - accuracy: 0.9980 - val_loss: 0.1461 - val_accuracy: 0.9980 - 72ms/epoch - 5ms/step\n", "Epoch 199/400\n", "16/16 - 0s - loss: 0.1426 - accuracy: 0.9980 - val_loss: 0.1448 - val_accuracy: 0.9980 - 122ms/epoch - 8ms/step\n", "Epoch 200/400\n", "16/16 - 0s - loss: 0.1411 - accuracy: 0.9980 - val_loss: 0.1434 - val_accuracy: 0.9980 - 109ms/epoch - 7ms/step\n", "Epoch 201/400\n", "16/16 - 0s - loss: 0.1396 - accuracy: 0.9980 - val_loss: 0.1421 - val_accuracy: 0.9980 - 71ms/epoch - 4ms/step\n", "Epoch 202/400\n", "16/16 - 0s - loss: 0.1382 - accuracy: 0.9980 - val_loss: 0.1408 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 203/400\n", "16/16 - 0s - loss: 0.1368 - accuracy: 0.9980 - val_loss: 0.1395 - val_accuracy: 0.9980 - 111ms/epoch - 7ms/step\n", "Epoch 204/400\n", "16/16 - 0s - loss: 0.1353 - accuracy: 0.9980 - val_loss: 0.1382 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 205/400\n", "16/16 - 0s - loss: 0.1339 - accuracy: 0.9980 - val_loss: 0.1369 - val_accuracy: 0.9980 - 71ms/epoch - 4ms/step\n", "Epoch 206/400\n", "16/16 - 0s - loss: 0.1326 - accuracy: 0.9980 - val_loss: 0.1357 - val_accuracy: 0.9980 - 68ms/epoch - 4ms/step\n", "Epoch 207/400\n", "16/16 - 0s - loss: 0.1312 - accuracy: 0.9980 - val_loss: 0.1345 - val_accuracy: 0.9980 - 72ms/epoch - 5ms/step\n", "Epoch 208/400\n", "16/16 - 0s - loss: 0.1299 - accuracy: 0.9980 - val_loss: 0.1333 - val_accuracy: 0.9980 - 73ms/epoch - 5ms/step\n", "Epoch 209/400\n", "16/16 - 0s - loss: 0.1286 - accuracy: 0.9980 - val_loss: 0.1321 - val_accuracy: 0.9980 - 83ms/epoch - 5ms/step\n", "Epoch 210/400\n", "16/16 - 0s - loss: 0.1272 - accuracy: 0.9980 - val_loss: 0.1309 - val_accuracy: 0.9980 - 78ms/epoch - 5ms/step\n", "Epoch 211/400\n", "16/16 - 0s - loss: 0.1259 - accuracy: 0.9980 - val_loss: 0.1298 - val_accuracy: 0.9980 - 112ms/epoch - 7ms/step\n", "Epoch 212/400\n", "16/16 - 0s - loss: 0.1246 - accuracy: 0.9980 - val_loss: 0.1287 - val_accuracy: 1.0000 - 112ms/epoch - 7ms/step\n", "Epoch 213/400\n", "16/16 - 0s - loss: 0.1233 - accuracy: 0.9980 - val_loss: 0.1276 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 214/400\n", "16/16 - 0s - loss: 0.1221 - accuracy: 0.9980 - val_loss: 0.1265 - val_accuracy: 1.0000 - 68ms/epoch - 4ms/step\n", "Epoch 215/400\n", "16/16 - 0s - loss: 0.1209 - accuracy: 0.9980 - val_loss: 0.1255 - val_accuracy: 1.0000 - 68ms/epoch - 4ms/step\n", "Epoch 216/400\n", "16/16 - 0s - loss: 0.1197 - accuracy: 0.9980 - val_loss: 0.1245 - val_accuracy: 1.0000 - 67ms/epoch - 4ms/step\n", "Epoch 217/400\n", "16/16 - 0s - loss: 0.1185 - accuracy: 0.9980 - val_loss: 0.1234 - val_accuracy: 1.0000 - 64ms/epoch - 4ms/step\n", "Epoch 218/400\n", "16/16 - 0s - loss: 0.1174 - accuracy: 0.9980 - val_loss: 0.1224 - val_accuracy: 1.0000 - 66ms/epoch - 4ms/step\n", "Epoch 219/400\n", "16/16 - 0s - loss: 0.1162 - accuracy: 0.9980 - val_loss: 0.1214 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 220/400\n", "16/16 - 0s - loss: 0.1151 - accuracy: 0.9980 - val_loss: 0.1204 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 221/400\n", "16/16 - 0s - loss: 0.1140 - accuracy: 0.9980 - val_loss: 0.1194 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 222/400\n", "16/16 - 0s - loss: 0.1129 - accuracy: 0.9980 - val_loss: 0.1185 - val_accuracy: 1.0000 - 77ms/epoch - 5ms/step\n", "Epoch 223/400\n", "16/16 - 0s - loss: 0.1118 - accuracy: 0.9980 - val_loss: 0.1175 - val_accuracy: 1.0000 - 72ms/epoch - 4ms/step\n", "Epoch 224/400\n", "16/16 - 0s - loss: 0.1107 - accuracy: 0.9980 - val_loss: 0.1165 - val_accuracy: 1.0000 - 67ms/epoch - 4ms/step\n", "Epoch 225/400\n", "16/16 - 0s - loss: 0.1097 - accuracy: 0.9980 - val_loss: 0.1156 - val_accuracy: 1.0000 - 65ms/epoch - 4ms/step\n", "Epoch 226/400\n", "16/16 - 0s - loss: 0.1086 - accuracy: 0.9980 - val_loss: 0.1147 - val_accuracy: 1.0000 - 69ms/epoch - 4ms/step\n", "Epoch 227/400\n", "16/16 - 0s - loss: 0.1077 - accuracy: 0.9980 - val_loss: 0.1137 - val_accuracy: 1.0000 - 66ms/epoch - 4ms/step\n", "Epoch 228/400\n", "16/16 - 0s - loss: 0.1067 - accuracy: 0.9980 - val_loss: 0.1128 - val_accuracy: 1.0000 - 69ms/epoch - 4ms/step\n", "Epoch 229/400\n", "16/16 - 0s - loss: 0.1057 - accuracy: 0.9980 - val_loss: 0.1120 - val_accuracy: 1.0000 - 68ms/epoch - 4ms/step\n", "Epoch 230/400\n", "16/16 - 0s - loss: 0.1047 - accuracy: 0.9980 - val_loss: 0.1111 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 231/400\n", "16/16 - 0s - loss: 0.1038 - accuracy: 0.9980 - val_loss: 0.1102 - val_accuracy: 1.0000 - 110ms/epoch - 7ms/step\n", "Epoch 232/400\n", "16/16 - 0s - loss: 0.1028 - accuracy: 1.0000 - val_loss: 0.1093 - val_accuracy: 1.0000 - 109ms/epoch - 7ms/step\n", "Epoch 233/400\n", "16/16 - 0s - loss: 0.1019 - accuracy: 1.0000 - val_loss: 0.1085 - val_accuracy: 1.0000 - 118ms/epoch - 7ms/step\n", "Epoch 234/400\n", "16/16 - 0s - loss: 0.1010 - accuracy: 1.0000 - val_loss: 0.1076 - val_accuracy: 1.0000 - 74ms/epoch - 5ms/step\n", "Epoch 235/400\n", "16/16 - 0s - loss: 0.1001 - accuracy: 1.0000 - val_loss: 0.1068 - val_accuracy: 1.0000 - 118ms/epoch - 7ms/step\n", "Epoch 236/400\n", "16/16 - 0s - loss: 0.0991 - accuracy: 1.0000 - val_loss: 0.1060 - val_accuracy: 1.0000 - 107ms/epoch - 7ms/step\n", "Epoch 237/400\n", "16/16 - 0s - loss: 0.0982 - accuracy: 1.0000 - val_loss: 0.1052 - val_accuracy: 1.0000 - 69ms/epoch - 4ms/step\n", "Epoch 238/400\n", "16/16 - 0s - loss: 0.0973 - accuracy: 1.0000 - val_loss: 0.1044 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 239/400\n", "16/16 - 0s - loss: 0.0965 - accuracy: 1.0000 - val_loss: 0.1036 - val_accuracy: 1.0000 - 114ms/epoch - 7ms/step\n", "Epoch 240/400\n", "16/16 - 0s - loss: 0.0956 - accuracy: 1.0000 - val_loss: 0.1029 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 241/400\n", "16/16 - 0s - loss: 0.0948 - accuracy: 1.0000 - val_loss: 0.1021 - val_accuracy: 1.0000 - 65ms/epoch - 4ms/step\n", "Epoch 242/400\n", "16/16 - 0s - loss: 0.0939 - accuracy: 1.0000 - val_loss: 0.1014 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 243/400\n", "16/16 - 0s - loss: 0.0931 - accuracy: 1.0000 - val_loss: 0.1006 - val_accuracy: 1.0000 - 112ms/epoch - 7ms/step\n", "Epoch 244/400\n", "16/16 - 0s - loss: 0.0923 - accuracy: 1.0000 - val_loss: 0.0998 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 245/400\n", "16/16 - 0s - loss: 0.0915 - accuracy: 1.0000 - val_loss: 0.0991 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 246/400\n", "16/16 - 0s - loss: 0.0907 - accuracy: 1.0000 - val_loss: 0.0984 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 247/400\n", "16/16 - 0s - loss: 0.0899 - accuracy: 1.0000 - val_loss: 0.0976 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 248/400\n", "16/16 - 0s - loss: 0.0892 - accuracy: 1.0000 - val_loss: 0.0969 - val_accuracy: 1.0000 - 108ms/epoch - 7ms/step\n", "Epoch 249/400\n", "16/16 - 0s - loss: 0.0884 - accuracy: 1.0000 - val_loss: 0.0962 - val_accuracy: 1.0000 - 69ms/epoch - 4ms/step\n", "Epoch 250/400\n", "16/16 - 0s - loss: 0.0877 - accuracy: 1.0000 - val_loss: 0.0955 - val_accuracy: 1.0000 - 69ms/epoch - 4ms/step\n", "Epoch 251/400\n", "16/16 - 0s - loss: 0.0869 - accuracy: 1.0000 - val_loss: 0.0948 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 252/400\n", "16/16 - 0s - loss: 0.0862 - accuracy: 1.0000 - val_loss: 0.0941 - val_accuracy: 1.0000 - 67ms/epoch - 4ms/step\n", "Epoch 253/400\n", "16/16 - 0s - loss: 0.0854 - accuracy: 1.0000 - val_loss: 0.0934 - val_accuracy: 1.0000 - 106ms/epoch - 7ms/step\n", "Epoch 254/400\n", "16/16 - 0s - loss: 0.0847 - accuracy: 1.0000 - val_loss: 0.0927 - val_accuracy: 1.0000 - 115ms/epoch - 7ms/step\n", "Epoch 255/400\n", "16/16 - 0s - loss: 0.0840 - accuracy: 1.0000 - val_loss: 0.0921 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 256/400\n", "16/16 - 0s - loss: 0.0833 - accuracy: 1.0000 - val_loss: 0.0914 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 257/400\n", "16/16 - 0s - loss: 0.0826 - accuracy: 1.0000 - val_loss: 0.0908 - val_accuracy: 1.0000 - 84ms/epoch - 5ms/step\n", "Epoch 258/400\n", "16/16 - 0s - loss: 0.0819 - accuracy: 1.0000 - val_loss: 0.0902 - val_accuracy: 1.0000 - 116ms/epoch - 7ms/step\n", "Epoch 259/400\n", "16/16 - 0s - loss: 0.0812 - accuracy: 1.0000 - val_loss: 0.0895 - val_accuracy: 1.0000 - 114ms/epoch - 7ms/step\n", "Epoch 260/400\n", "16/16 - 0s - loss: 0.0806 - accuracy: 1.0000 - val_loss: 0.0889 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 261/400\n", "16/16 - 0s - loss: 0.0799 - accuracy: 1.0000 - val_loss: 0.0883 - val_accuracy: 1.0000 - 65ms/epoch - 4ms/step\n", "Epoch 262/400\n", "16/16 - 0s - loss: 0.0793 - accuracy: 1.0000 - val_loss: 0.0877 - val_accuracy: 1.0000 - 68ms/epoch - 4ms/step\n", "Epoch 263/400\n", "16/16 - 0s - loss: 0.0786 - accuracy: 1.0000 - val_loss: 0.0871 - val_accuracy: 1.0000 - 66ms/epoch - 4ms/step\n", "Epoch 264/400\n", "16/16 - 0s - loss: 0.0780 - accuracy: 1.0000 - val_loss: 0.0865 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 265/400\n", "16/16 - 0s - loss: 0.0774 - accuracy: 1.0000 - val_loss: 0.0859 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 266/400\n", "16/16 - 0s - loss: 0.0768 - accuracy: 1.0000 - val_loss: 0.0853 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 267/400\n", "16/16 - 0s - loss: 0.0762 - accuracy: 1.0000 - val_loss: 0.0848 - val_accuracy: 1.0000 - 68ms/epoch - 4ms/step\n", "Epoch 268/400\n", "16/16 - 0s - loss: 0.0756 - accuracy: 1.0000 - val_loss: 0.0842 - val_accuracy: 1.0000 - 68ms/epoch - 4ms/step\n", "Epoch 269/400\n", "16/16 - 0s - loss: 0.0750 - accuracy: 1.0000 - val_loss: 0.0836 - val_accuracy: 1.0000 - 78ms/epoch - 5ms/step\n", "Epoch 270/400\n", "16/16 - 0s - loss: 0.0744 - accuracy: 1.0000 - val_loss: 0.0831 - val_accuracy: 1.0000 - 69ms/epoch - 4ms/step\n", "Epoch 271/400\n", "16/16 - 0s - loss: 0.0738 - accuracy: 1.0000 - val_loss: 0.0825 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 272/400\n", "16/16 - 0s - loss: 0.0732 - accuracy: 1.0000 - val_loss: 0.0820 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 273/400\n", "16/16 - 0s - loss: 0.0726 - accuracy: 1.0000 - val_loss: 0.0814 - val_accuracy: 1.0000 - 110ms/epoch - 7ms/step\n", "Epoch 274/400\n", "16/16 - 0s - loss: 0.0721 - accuracy: 1.0000 - val_loss: 0.0809 - val_accuracy: 1.0000 - 67ms/epoch - 4ms/step\n", "Epoch 275/400\n", "16/16 - 0s - loss: 0.0715 - accuracy: 1.0000 - val_loss: 0.0804 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 276/400\n", "16/16 - 0s - loss: 0.0709 - accuracy: 1.0000 - val_loss: 0.0798 - val_accuracy: 1.0000 - 68ms/epoch - 4ms/step\n", "Epoch 277/400\n", "16/16 - 0s - loss: 0.0703 - accuracy: 1.0000 - val_loss: 0.0793 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 278/400\n", "16/16 - 0s - loss: 0.0698 - accuracy: 1.0000 - val_loss: 0.0787 - val_accuracy: 1.0000 - 68ms/epoch - 4ms/step\n", "Epoch 279/400\n", "16/16 - 0s - loss: 0.0692 - accuracy: 1.0000 - val_loss: 0.0782 - val_accuracy: 1.0000 - 116ms/epoch - 7ms/step\n", "Epoch 280/400\n", "16/16 - 0s - loss: 0.0687 - accuracy: 1.0000 - val_loss: 0.0777 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 281/400\n", "16/16 - 0s - loss: 0.0681 - accuracy: 1.0000 - val_loss: 0.0771 - val_accuracy: 1.0000 - 110ms/epoch - 7ms/step\n", "Epoch 282/400\n", "16/16 - 0s - loss: 0.0676 - accuracy: 1.0000 - val_loss: 0.0766 - val_accuracy: 1.0000 - 111ms/epoch - 7ms/step\n", "Epoch 283/400\n", "16/16 - 0s - loss: 0.0671 - accuracy: 1.0000 - val_loss: 0.0761 - val_accuracy: 1.0000 - 110ms/epoch - 7ms/step\n", "Epoch 284/400\n", "16/16 - 0s - loss: 0.0665 - accuracy: 1.0000 - val_loss: 0.0756 - val_accuracy: 1.0000 - 67ms/epoch - 4ms/step\n", "Epoch 285/400\n", "16/16 - 0s - loss: 0.0660 - accuracy: 1.0000 - val_loss: 0.0752 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 286/400\n", "16/16 - 0s - loss: 0.0655 - accuracy: 1.0000 - val_loss: 0.0746 - val_accuracy: 1.0000 - 72ms/epoch - 4ms/step\n", "Epoch 287/400\n", "16/16 - 0s - loss: 0.0650 - accuracy: 1.0000 - val_loss: 0.0741 - val_accuracy: 1.0000 - 66ms/epoch - 4ms/step\n", "Epoch 288/400\n", "16/16 - 0s - loss: 0.0645 - accuracy: 1.0000 - val_loss: 0.0737 - val_accuracy: 1.0000 - 121ms/epoch - 8ms/step\n", "Epoch 289/400\n", "16/16 - 0s - loss: 0.0640 - accuracy: 1.0000 - val_loss: 0.0732 - val_accuracy: 1.0000 - 113ms/epoch - 7ms/step\n", "Epoch 290/400\n", "16/16 - 0s - loss: 0.0635 - accuracy: 1.0000 - val_loss: 0.0727 - val_accuracy: 1.0000 - 68ms/epoch - 4ms/step\n", "Epoch 291/400\n", "16/16 - 0s - loss: 0.0630 - accuracy: 1.0000 - val_loss: 0.0722 - val_accuracy: 1.0000 - 69ms/epoch - 4ms/step\n", "Epoch 292/400\n", "16/16 - 0s - loss: 0.0625 - accuracy: 1.0000 - val_loss: 0.0717 - val_accuracy: 1.0000 - 81ms/epoch - 5ms/step\n", "Epoch 293/400\n", "16/16 - 0s - loss: 0.0620 - accuracy: 1.0000 - val_loss: 0.0712 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 294/400\n", "16/16 - 0s - loss: 0.0616 - accuracy: 1.0000 - val_loss: 0.0708 - val_accuracy: 1.0000 - 69ms/epoch - 4ms/step\n", "Epoch 295/400\n", "16/16 - 0s - loss: 0.0611 - accuracy: 1.0000 - val_loss: 0.0703 - val_accuracy: 1.0000 - 68ms/epoch - 4ms/step\n", "Epoch 296/400\n", "16/16 - 0s - loss: 0.0606 - accuracy: 1.0000 - val_loss: 0.0699 - val_accuracy: 1.0000 - 111ms/epoch - 7ms/step\n", "Epoch 297/400\n", "16/16 - 0s - loss: 0.0601 - accuracy: 1.0000 - val_loss: 0.0694 - val_accuracy: 1.0000 - 68ms/epoch - 4ms/step\n", "Epoch 298/400\n", "16/16 - 0s - loss: 0.0597 - accuracy: 1.0000 - val_loss: 0.0690 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 299/400\n", "16/16 - 0s - loss: 0.0592 - accuracy: 1.0000 - val_loss: 0.0685 - val_accuracy: 1.0000 - 110ms/epoch - 7ms/step\n", "Epoch 300/400\n", "16/16 - 0s - loss: 0.0588 - accuracy: 1.0000 - val_loss: 0.0680 - val_accuracy: 1.0000 - 113ms/epoch - 7ms/step\n", "Epoch 301/400\n", "16/16 - 0s - loss: 0.0583 - accuracy: 1.0000 - val_loss: 0.0676 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 302/400\n", "16/16 - 0s - loss: 0.0579 - accuracy: 1.0000 - val_loss: 0.0671 - val_accuracy: 1.0000 - 72ms/epoch - 4ms/step\n", "Epoch 303/400\n", "16/16 - 0s - loss: 0.0574 - accuracy: 1.0000 - val_loss: 0.0667 - val_accuracy: 1.0000 - 119ms/epoch - 7ms/step\n", "Epoch 304/400\n", "16/16 - 0s - loss: 0.0570 - accuracy: 1.0000 - val_loss: 0.0663 - val_accuracy: 1.0000 - 108ms/epoch - 7ms/step\n", "Epoch 305/400\n", "16/16 - 0s - loss: 0.0566 - accuracy: 1.0000 - val_loss: 0.0659 - val_accuracy: 1.0000 - 80ms/epoch - 5ms/step\n", "Epoch 306/400\n", "16/16 - 0s - loss: 0.0562 - accuracy: 1.0000 - val_loss: 0.0655 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 307/400\n", "16/16 - 0s - loss: 0.0558 - accuracy: 1.0000 - val_loss: 0.0651 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 308/400\n", "16/16 - 0s - loss: 0.0553 - accuracy: 1.0000 - val_loss: 0.0647 - val_accuracy: 1.0000 - 111ms/epoch - 7ms/step\n", "Epoch 309/400\n", "16/16 - 0s - loss: 0.0549 - accuracy: 1.0000 - val_loss: 0.0642 - val_accuracy: 1.0000 - 76ms/epoch - 5ms/step\n", "Epoch 310/400\n", "16/16 - 0s - loss: 0.0545 - accuracy: 1.0000 - val_loss: 0.0638 - val_accuracy: 1.0000 - 121ms/epoch - 8ms/step\n", "Epoch 311/400\n", "16/16 - 0s - loss: 0.0541 - accuracy: 1.0000 - val_loss: 0.0634 - val_accuracy: 1.0000 - 114ms/epoch - 7ms/step\n", "Epoch 312/400\n", "16/16 - 0s - loss: 0.0537 - accuracy: 1.0000 - val_loss: 0.0630 - val_accuracy: 1.0000 - 66ms/epoch - 4ms/step\n", "Epoch 313/400\n", "16/16 - 0s - loss: 0.0534 - accuracy: 1.0000 - val_loss: 0.0626 - val_accuracy: 1.0000 - 68ms/epoch - 4ms/step\n", "Epoch 314/400\n", "16/16 - 0s - loss: 0.0530 - accuracy: 1.0000 - val_loss: 0.0623 - val_accuracy: 1.0000 - 112ms/epoch - 7ms/step\n", "Epoch 315/400\n", "16/16 - 0s - loss: 0.0526 - accuracy: 1.0000 - val_loss: 0.0619 - val_accuracy: 1.0000 - 68ms/epoch - 4ms/step\n", "Epoch 316/400\n", "16/16 - 0s - loss: 0.0522 - accuracy: 1.0000 - val_loss: 0.0615 - val_accuracy: 1.0000 - 72ms/epoch - 4ms/step\n", "Epoch 317/400\n", "16/16 - 0s - loss: 0.0518 - accuracy: 1.0000 - val_loss: 0.0612 - val_accuracy: 1.0000 - 67ms/epoch - 4ms/step\n", "Epoch 318/400\n", "16/16 - 0s - loss: 0.0515 - accuracy: 1.0000 - val_loss: 0.0608 - val_accuracy: 1.0000 - 113ms/epoch - 7ms/step\n", "Epoch 319/400\n", "16/16 - 0s - loss: 0.0511 - accuracy: 1.0000 - val_loss: 0.0605 - val_accuracy: 1.0000 - 66ms/epoch - 4ms/step\n", "Epoch 320/400\n", "16/16 - 0s - loss: 0.0508 - accuracy: 1.0000 - val_loss: 0.0601 - val_accuracy: 1.0000 - 74ms/epoch - 5ms/step\n", "Epoch 321/400\n", "16/16 - 0s - loss: 0.0504 - accuracy: 1.0000 - val_loss: 0.0597 - val_accuracy: 0.9980 - 67ms/epoch - 4ms/step\n", "Epoch 322/400\n", "16/16 - 0s - loss: 0.0500 - accuracy: 1.0000 - val_loss: 0.0594 - val_accuracy: 0.9980 - 118ms/epoch - 7ms/step\n", "Epoch 323/400\n", "16/16 - 0s - loss: 0.0497 - accuracy: 1.0000 - val_loss: 0.0591 - val_accuracy: 0.9980 - 67ms/epoch - 4ms/step\n", "Epoch 324/400\n", "16/16 - 0s - loss: 0.0494 - accuracy: 1.0000 - val_loss: 0.0587 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 325/400\n", "16/16 - 0s - loss: 0.0490 - accuracy: 1.0000 - val_loss: 0.0583 - val_accuracy: 0.9980 - 113ms/epoch - 7ms/step\n", "Epoch 326/400\n", "16/16 - 0s - loss: 0.0487 - accuracy: 1.0000 - val_loss: 0.0580 - val_accuracy: 0.9980 - 83ms/epoch - 5ms/step\n", "Epoch 327/400\n", "16/16 - 0s - loss: 0.0483 - accuracy: 1.0000 - val_loss: 0.0577 - val_accuracy: 0.9980 - 76ms/epoch - 5ms/step\n", "Epoch 328/400\n", "16/16 - 0s - loss: 0.0480 - accuracy: 1.0000 - val_loss: 0.0573 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 329/400\n", "16/16 - 0s - loss: 0.0477 - accuracy: 1.0000 - val_loss: 0.0570 - val_accuracy: 0.9980 - 72ms/epoch - 5ms/step\n", "Epoch 330/400\n", "16/16 - 0s - loss: 0.0474 - accuracy: 1.0000 - val_loss: 0.0567 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 331/400\n", "16/16 - 0s - loss: 0.0470 - accuracy: 1.0000 - val_loss: 0.0563 - val_accuracy: 0.9980 - 109ms/epoch - 7ms/step\n", "Epoch 332/400\n", "16/16 - 0s - loss: 0.0467 - accuracy: 1.0000 - val_loss: 0.0561 - val_accuracy: 0.9980 - 113ms/epoch - 7ms/step\n", "Epoch 333/400\n", "16/16 - 0s - loss: 0.0464 - accuracy: 1.0000 - val_loss: 0.0557 - val_accuracy: 0.9980 - 67ms/epoch - 4ms/step\n", "Epoch 334/400\n", "16/16 - 0s - loss: 0.0461 - accuracy: 1.0000 - val_loss: 0.0554 - val_accuracy: 0.9980 - 114ms/epoch - 7ms/step\n", "Epoch 335/400\n", "16/16 - 0s - loss: 0.0458 - accuracy: 1.0000 - val_loss: 0.0551 - val_accuracy: 0.9980 - 94ms/epoch - 6ms/step\n", "Epoch 336/400\n", "16/16 - 0s - loss: 0.0455 - accuracy: 1.0000 - val_loss: 0.0547 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 337/400\n", "16/16 - 0s - loss: 0.0452 - accuracy: 1.0000 - val_loss: 0.0544 - val_accuracy: 0.9980 - 73ms/epoch - 5ms/step\n", "Epoch 338/400\n", "16/16 - 0s - loss: 0.0449 - accuracy: 1.0000 - val_loss: 0.0541 - val_accuracy: 0.9980 - 71ms/epoch - 4ms/step\n", "Epoch 339/400\n", "16/16 - 0s - loss: 0.0446 - accuracy: 1.0000 - val_loss: 0.0538 - val_accuracy: 0.9980 - 66ms/epoch - 4ms/step\n", "Epoch 340/400\n", "16/16 - 0s - loss: 0.0443 - accuracy: 1.0000 - val_loss: 0.0535 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 341/400\n", "16/16 - 0s - loss: 0.0440 - accuracy: 1.0000 - val_loss: 0.0532 - val_accuracy: 0.9980 - 113ms/epoch - 7ms/step\n", "Epoch 342/400\n", "16/16 - 0s - loss: 0.0437 - accuracy: 1.0000 - val_loss: 0.0529 - val_accuracy: 0.9980 - 114ms/epoch - 7ms/step\n", "Epoch 343/400\n", "16/16 - 0s - loss: 0.0434 - accuracy: 1.0000 - val_loss: 0.0527 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 344/400\n", "16/16 - 0s - loss: 0.0431 - accuracy: 1.0000 - val_loss: 0.0524 - val_accuracy: 0.9980 - 68ms/epoch - 4ms/step\n", "Epoch 345/400\n", "16/16 - 0s - loss: 0.0428 - accuracy: 1.0000 - val_loss: 0.0521 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 346/400\n", "16/16 - 0s - loss: 0.0426 - accuracy: 1.0000 - val_loss: 0.0518 - val_accuracy: 0.9980 - 66ms/epoch - 4ms/step\n", "Epoch 347/400\n", "16/16 - 0s - loss: 0.0423 - accuracy: 1.0000 - val_loss: 0.0515 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 348/400\n", "16/16 - 0s - loss: 0.0420 - accuracy: 1.0000 - val_loss: 0.0512 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 349/400\n", "16/16 - 0s - loss: 0.0417 - accuracy: 1.0000 - val_loss: 0.0509 - val_accuracy: 0.9980 - 77ms/epoch - 5ms/step\n", "Epoch 350/400\n", "16/16 - 0s - loss: 0.0415 - accuracy: 1.0000 - val_loss: 0.0506 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 351/400\n", "16/16 - 0s - loss: 0.0412 - accuracy: 1.0000 - val_loss: 0.0503 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 352/400\n", "16/16 - 0s - loss: 0.0409 - accuracy: 1.0000 - val_loss: 0.0500 - val_accuracy: 0.9980 - 68ms/epoch - 4ms/step\n", "Epoch 353/400\n", "16/16 - 0s - loss: 0.0407 - accuracy: 1.0000 - val_loss: 0.0498 - val_accuracy: 0.9980 - 72ms/epoch - 4ms/step\n", "Epoch 354/400\n", "16/16 - 0s - loss: 0.0404 - accuracy: 1.0000 - val_loss: 0.0495 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 355/400\n", "16/16 - 0s - loss: 0.0401 - accuracy: 1.0000 - val_loss: 0.0493 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 356/400\n", "16/16 - 0s - loss: 0.0399 - accuracy: 1.0000 - val_loss: 0.0490 - val_accuracy: 0.9980 - 110ms/epoch - 7ms/step\n", "Epoch 357/400\n", "16/16 - 0s - loss: 0.0396 - accuracy: 1.0000 - val_loss: 0.0488 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 358/400\n", "16/16 - 0s - loss: 0.0394 - accuracy: 1.0000 - val_loss: 0.0485 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 359/400\n", "16/16 - 0s - loss: 0.0391 - accuracy: 1.0000 - val_loss: 0.0482 - val_accuracy: 0.9980 - 112ms/epoch - 7ms/step\n", "Epoch 360/400\n", "16/16 - 0s - loss: 0.0389 - accuracy: 1.0000 - val_loss: 0.0479 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 361/400\n", "16/16 - 0s - loss: 0.0386 - accuracy: 1.0000 - val_loss: 0.0477 - val_accuracy: 0.9980 - 76ms/epoch - 5ms/step\n", "Epoch 362/400\n", "16/16 - 0s - loss: 0.0384 - accuracy: 1.0000 - val_loss: 0.0475 - val_accuracy: 0.9980 - 126ms/epoch - 8ms/step\n", "Epoch 363/400\n", "16/16 - 0s - loss: 0.0381 - accuracy: 1.0000 - val_loss: 0.0472 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 364/400\n", "16/16 - 0s - loss: 0.0379 - accuracy: 1.0000 - val_loss: 0.0469 - val_accuracy: 0.9980 - 112ms/epoch - 7ms/step\n", "Epoch 365/400\n", "16/16 - 0s - loss: 0.0377 - accuracy: 1.0000 - val_loss: 0.0467 - val_accuracy: 0.9980 - 117ms/epoch - 7ms/step\n", "Epoch 366/400\n", "16/16 - 0s - loss: 0.0374 - accuracy: 1.0000 - val_loss: 0.0465 - val_accuracy: 0.9980 - 112ms/epoch - 7ms/step\n", "Epoch 367/400\n", "16/16 - 0s - loss: 0.0372 - accuracy: 1.0000 - val_loss: 0.0462 - val_accuracy: 0.9980 - 72ms/epoch - 4ms/step\n", "Epoch 368/400\n", "16/16 - 0s - loss: 0.0370 - accuracy: 1.0000 - val_loss: 0.0460 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 369/400\n", "16/16 - 0s - loss: 0.0367 - accuracy: 1.0000 - val_loss: 0.0457 - val_accuracy: 0.9980 - 111ms/epoch - 7ms/step\n", "Epoch 370/400\n", "16/16 - 0s - loss: 0.0365 - accuracy: 1.0000 - val_loss: 0.0455 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 371/400\n", "16/16 - 0s - loss: 0.0363 - accuracy: 1.0000 - val_loss: 0.0453 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 372/400\n", "16/16 - 0s - loss: 0.0360 - accuracy: 1.0000 - val_loss: 0.0450 - val_accuracy: 0.9980 - 118ms/epoch - 7ms/step\n", "Epoch 373/400\n", "16/16 - 0s - loss: 0.0358 - accuracy: 1.0000 - val_loss: 0.0448 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 374/400\n", "16/16 - 0s - loss: 0.0356 - accuracy: 1.0000 - val_loss: 0.0445 - val_accuracy: 0.9980 - 112ms/epoch - 7ms/step\n", "Epoch 375/400\n", "16/16 - 0s - loss: 0.0354 - accuracy: 1.0000 - val_loss: 0.0442 - val_accuracy: 0.9980 - 72ms/epoch - 5ms/step\n", "Epoch 376/400\n", "16/16 - 0s - loss: 0.0351 - accuracy: 1.0000 - val_loss: 0.0440 - val_accuracy: 0.9980 - 72ms/epoch - 5ms/step\n", "Epoch 377/400\n", "16/16 - 0s - loss: 0.0349 - accuracy: 1.0000 - val_loss: 0.0437 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 378/400\n", "16/16 - 0s - loss: 0.0347 - accuracy: 1.0000 - val_loss: 0.0434 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 379/400\n", "16/16 - 0s - loss: 0.0344 - accuracy: 1.0000 - val_loss: 0.0431 - val_accuracy: 0.9980 - 116ms/epoch - 7ms/step\n", "Epoch 380/400\n", "16/16 - 0s - loss: 0.0342 - accuracy: 1.0000 - val_loss: 0.0428 - val_accuracy: 0.9980 - 72ms/epoch - 4ms/step\n", "Epoch 381/400\n", "16/16 - 0s - loss: 0.0339 - accuracy: 1.0000 - val_loss: 0.0425 - val_accuracy: 0.9980 - 111ms/epoch - 7ms/step\n", "Epoch 382/400\n", "16/16 - 0s - loss: 0.0337 - accuracy: 1.0000 - val_loss: 0.0422 - val_accuracy: 0.9980 - 109ms/epoch - 7ms/step\n", "Epoch 383/400\n", "16/16 - 0s - loss: 0.0335 - accuracy: 1.0000 - val_loss: 0.0420 - val_accuracy: 0.9980 - 78ms/epoch - 5ms/step\n", "Epoch 384/400\n", "16/16 - 0s - loss: 0.0333 - accuracy: 1.0000 - val_loss: 0.0417 - val_accuracy: 0.9980 - 71ms/epoch - 4ms/step\n", "Epoch 385/400\n", "16/16 - 0s - loss: 0.0330 - accuracy: 1.0000 - val_loss: 0.0414 - val_accuracy: 0.9980 - 72ms/epoch - 5ms/step\n", "Epoch 386/400\n", "16/16 - 0s - loss: 0.0328 - accuracy: 1.0000 - val_loss: 0.0412 - val_accuracy: 0.9980 - 76ms/epoch - 5ms/step\n", "Epoch 387/400\n", "16/16 - 0s - loss: 0.0326 - accuracy: 1.0000 - val_loss: 0.0409 - val_accuracy: 0.9980 - 110ms/epoch - 7ms/step\n", "Epoch 388/400\n", "16/16 - 0s - loss: 0.0324 - accuracy: 1.0000 - val_loss: 0.0407 - val_accuracy: 0.9980 - 114ms/epoch - 7ms/step\n", "Epoch 389/400\n", "16/16 - 0s - loss: 0.0322 - accuracy: 1.0000 - val_loss: 0.0404 - val_accuracy: 0.9980 - 79ms/epoch - 5ms/step\n", "Epoch 390/400\n", "16/16 - 0s - loss: 0.0320 - accuracy: 1.0000 - val_loss: 0.0402 - val_accuracy: 0.9980 - 71ms/epoch - 4ms/step\n", "Epoch 391/400\n", "16/16 - 0s - loss: 0.0318 - accuracy: 1.0000 - val_loss: 0.0400 - val_accuracy: 0.9980 - 68ms/epoch - 4ms/step\n", "Epoch 392/400\n", "16/16 - 0s - loss: 0.0316 - accuracy: 1.0000 - val_loss: 0.0397 - val_accuracy: 0.9980 - 114ms/epoch - 7ms/step\n", "Epoch 393/400\n", "16/16 - 0s - loss: 0.0314 - accuracy: 1.0000 - val_loss: 0.0395 - val_accuracy: 0.9980 - 71ms/epoch - 4ms/step\n", "Epoch 394/400\n", "16/16 - 0s - loss: 0.0312 - accuracy: 1.0000 - val_loss: 0.0393 - val_accuracy: 0.9980 - 89ms/epoch - 6ms/step\n", "Epoch 395/400\n", "16/16 - 0s - loss: 0.0310 - accuracy: 1.0000 - val_loss: 0.0391 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 396/400\n", "16/16 - 0s - loss: 0.0308 - accuracy: 1.0000 - val_loss: 0.0388 - val_accuracy: 0.9980 - 72ms/epoch - 4ms/step\n", "Epoch 397/400\n", "16/16 - 0s - loss: 0.0306 - accuracy: 1.0000 - val_loss: 0.0386 - val_accuracy: 0.9980 - 80ms/epoch - 5ms/step\n", "Epoch 398/400\n", "16/16 - 0s - loss: 0.0304 - accuracy: 1.0000 - val_loss: 0.0384 - val_accuracy: 0.9980 - 72ms/epoch - 5ms/step\n", "Epoch 399/400\n", "16/16 - 0s - loss: 0.0302 - accuracy: 1.0000 - val_loss: 0.0382 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 400/400\n", "16/16 - 0s - loss: 0.0300 - accuracy: 1.0000 - val_loss: 0.0379 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "第2个弱分类器训练完毕\n", "Epoch 1/400\n", "16/16 - 1s - loss: 0.5794 - accuracy: 0.7400 - val_loss: 0.6263 - val_accuracy: 0.6960 - 812ms/epoch - 51ms/step\n", "Epoch 2/400\n", "16/16 - 0s - loss: 0.5671 - accuracy: 0.7400 - val_loss: 0.6141 - val_accuracy: 0.7040 - 71ms/epoch - 4ms/step\n", "Epoch 3/400\n", "16/16 - 0s - loss: 0.5552 - accuracy: 0.7440 - val_loss: 0.6023 - val_accuracy: 0.7040 - 122ms/epoch - 8ms/step\n", "Epoch 4/400\n", "16/16 - 0s - loss: 0.5438 - accuracy: 0.7440 - val_loss: 0.5907 - val_accuracy: 0.7040 - 67ms/epoch - 4ms/step\n", "Epoch 5/400\n", "16/16 - 0s - loss: 0.5325 - accuracy: 0.7460 - val_loss: 0.5796 - val_accuracy: 0.7080 - 87ms/epoch - 5ms/step\n", "Epoch 6/400\n", "16/16 - 0s - loss: 0.5217 - accuracy: 0.7460 - val_loss: 0.5685 - val_accuracy: 0.7100 - 72ms/epoch - 4ms/step\n", "Epoch 7/400\n", "16/16 - 0s - loss: 0.5112 - accuracy: 0.7500 - val_loss: 0.5578 - val_accuracy: 0.7160 - 71ms/epoch - 4ms/step\n", "Epoch 8/400\n", "16/16 - 0s - loss: 0.5008 - accuracy: 0.7520 - val_loss: 0.5475 - val_accuracy: 0.7240 - 124ms/epoch - 8ms/step\n", "Epoch 9/400\n", "16/16 - 0s - loss: 0.4908 - accuracy: 0.7560 - val_loss: 0.5377 - val_accuracy: 0.7260 - 113ms/epoch - 7ms/step\n", "Epoch 10/400\n", "16/16 - 0s - loss: 0.4811 - accuracy: 0.7600 - val_loss: 0.5281 - val_accuracy: 0.7320 - 73ms/epoch - 5ms/step\n", "Epoch 11/400\n", "16/16 - 0s - loss: 0.4720 - accuracy: 0.7620 - val_loss: 0.5179 - val_accuracy: 0.7420 - 108ms/epoch - 7ms/step\n", "Epoch 12/400\n", "16/16 - 0s - loss: 0.4622 - accuracy: 0.7640 - val_loss: 0.5087 - val_accuracy: 0.7440 - 112ms/epoch - 7ms/step\n", "Epoch 13/400\n", "16/16 - 0s - loss: 0.4535 - accuracy: 0.7660 - val_loss: 0.4993 - val_accuracy: 0.7440 - 72ms/epoch - 5ms/step\n", "Epoch 14/400\n", "16/16 - 0s - loss: 0.4447 - accuracy: 0.7680 - val_loss: 0.4902 - val_accuracy: 0.7480 - 113ms/epoch - 7ms/step\n", "Epoch 15/400\n", "16/16 - 0s - loss: 0.4362 - accuracy: 0.7680 - val_loss: 0.4815 - val_accuracy: 0.7540 - 70ms/epoch - 4ms/step\n", "Epoch 16/400\n", "16/16 - 0s - loss: 0.4277 - accuracy: 0.7720 - val_loss: 0.4735 - val_accuracy: 0.7560 - 85ms/epoch - 5ms/step\n", "Epoch 17/400\n", "16/16 - 0s - loss: 0.4199 - accuracy: 0.7780 - val_loss: 0.4653 - val_accuracy: 0.7600 - 111ms/epoch - 7ms/step\n", "Epoch 18/400\n", "16/16 - 0s - loss: 0.4122 - accuracy: 0.7800 - val_loss: 0.4572 - val_accuracy: 0.7600 - 71ms/epoch - 4ms/step\n", "Epoch 19/400\n", "16/16 - 0s - loss: 0.4046 - accuracy: 0.7800 - val_loss: 0.4495 - val_accuracy: 0.7620 - 70ms/epoch - 4ms/step\n", "Epoch 20/400\n", "16/16 - 0s - loss: 0.3971 - accuracy: 0.7840 - val_loss: 0.4419 - val_accuracy: 0.7640 - 113ms/epoch - 7ms/step\n", "Epoch 21/400\n", "16/16 - 0s - loss: 0.3899 - accuracy: 0.7900 - val_loss: 0.4345 - val_accuracy: 0.7720 - 111ms/epoch - 7ms/step\n", "Epoch 22/400\n", "16/16 - 0s - loss: 0.3829 - accuracy: 0.7960 - val_loss: 0.4272 - val_accuracy: 0.7720 - 119ms/epoch - 7ms/step\n", "Epoch 23/400\n", "16/16 - 0s - loss: 0.3761 - accuracy: 0.7960 - val_loss: 0.4201 - val_accuracy: 0.7780 - 77ms/epoch - 5ms/step\n", "Epoch 24/400\n", "16/16 - 0s - loss: 0.3695 - accuracy: 0.8000 - val_loss: 0.4132 - val_accuracy: 0.7780 - 76ms/epoch - 5ms/step\n", "Epoch 25/400\n", "16/16 - 0s - loss: 0.3631 - accuracy: 0.8080 - val_loss: 0.4066 - val_accuracy: 0.7780 - 81ms/epoch - 5ms/step\n", "Epoch 26/400\n", "16/16 - 0s - loss: 0.3568 - accuracy: 0.8100 - val_loss: 0.4002 - val_accuracy: 0.7820 - 112ms/epoch - 7ms/step\n", "Epoch 27/400\n", "16/16 - 0s - loss: 0.3506 - accuracy: 0.8140 - val_loss: 0.3940 - val_accuracy: 0.7860 - 70ms/epoch - 4ms/step\n", "Epoch 28/400\n", "16/16 - 0s - loss: 0.3448 - accuracy: 0.8220 - val_loss: 0.3876 - val_accuracy: 0.7860 - 108ms/epoch - 7ms/step\n", "Epoch 29/400\n", "16/16 - 0s - loss: 0.3390 - accuracy: 0.8240 - val_loss: 0.3817 - val_accuracy: 0.7880 - 117ms/epoch - 7ms/step\n", "Epoch 30/400\n", "16/16 - 0s - loss: 0.3333 - accuracy: 0.8300 - val_loss: 0.3761 - val_accuracy: 0.8020 - 73ms/epoch - 5ms/step\n", "Epoch 31/400\n", "16/16 - 0s - loss: 0.3281 - accuracy: 0.8320 - val_loss: 0.3702 - val_accuracy: 0.8040 - 74ms/epoch - 5ms/step\n", "Epoch 32/400\n", "16/16 - 0s - loss: 0.3228 - accuracy: 0.8360 - val_loss: 0.3645 - val_accuracy: 0.8060 - 68ms/epoch - 4ms/step\n", "Epoch 33/400\n", "16/16 - 0s - loss: 0.3175 - accuracy: 0.8380 - val_loss: 0.3593 - val_accuracy: 0.8120 - 68ms/epoch - 4ms/step\n", "Epoch 34/400\n", "16/16 - 0s - loss: 0.3125 - accuracy: 0.8460 - val_loss: 0.3541 - val_accuracy: 0.8180 - 67ms/epoch - 4ms/step\n", "Epoch 35/400\n", "16/16 - 0s - loss: 0.3077 - accuracy: 0.8500 - val_loss: 0.3489 - val_accuracy: 0.8180 - 73ms/epoch - 5ms/step\n", "Epoch 36/400\n", "16/16 - 0s - loss: 0.3031 - accuracy: 0.8580 - val_loss: 0.3439 - val_accuracy: 0.8220 - 74ms/epoch - 5ms/step\n", "Epoch 37/400\n", "16/16 - 0s - loss: 0.2984 - accuracy: 0.8600 - val_loss: 0.3391 - val_accuracy: 0.8260 - 110ms/epoch - 7ms/step\n", "Epoch 38/400\n", "16/16 - 0s - loss: 0.2941 - accuracy: 0.8660 - val_loss: 0.3343 - val_accuracy: 0.8300 - 82ms/epoch - 5ms/step\n", "Epoch 39/400\n", "16/16 - 0s - loss: 0.2897 - accuracy: 0.8680 - val_loss: 0.3298 - val_accuracy: 0.8340 - 74ms/epoch - 5ms/step\n", "Epoch 40/400\n", "16/16 - 0s - loss: 0.2855 - accuracy: 0.8680 - val_loss: 0.3253 - val_accuracy: 0.8380 - 111ms/epoch - 7ms/step\n", "Epoch 41/400\n", "16/16 - 0s - loss: 0.2815 - accuracy: 0.8720 - val_loss: 0.3209 - val_accuracy: 0.8400 - 74ms/epoch - 5ms/step\n", "Epoch 42/400\n", "16/16 - 0s - loss: 0.2774 - accuracy: 0.8740 - val_loss: 0.3168 - val_accuracy: 0.8440 - 69ms/epoch - 4ms/step\n", "Epoch 43/400\n", "16/16 - 0s - loss: 0.2737 - accuracy: 0.8760 - val_loss: 0.3126 - val_accuracy: 0.8520 - 68ms/epoch - 4ms/step\n", "Epoch 44/400\n", "16/16 - 0s - loss: 0.2700 - accuracy: 0.8780 - val_loss: 0.3084 - val_accuracy: 0.8580 - 111ms/epoch - 7ms/step\n", "Epoch 45/400\n", "16/16 - 0s - loss: 0.2661 - accuracy: 0.8880 - val_loss: 0.3048 - val_accuracy: 0.8620 - 73ms/epoch - 5ms/step\n", "Epoch 46/400\n", "16/16 - 0s - loss: 0.2627 - accuracy: 0.8900 - val_loss: 0.3009 - val_accuracy: 0.8660 - 110ms/epoch - 7ms/step\n", "Epoch 47/400\n", "16/16 - 0s - loss: 0.2592 - accuracy: 0.8920 - val_loss: 0.2972 - val_accuracy: 0.8680 - 107ms/epoch - 7ms/step\n", "Epoch 48/400\n", "16/16 - 0s - loss: 0.2560 - accuracy: 0.8960 - val_loss: 0.2935 - val_accuracy: 0.8680 - 76ms/epoch - 5ms/step\n", "Epoch 49/400\n", "16/16 - 0s - loss: 0.2526 - accuracy: 0.8960 - val_loss: 0.2898 - val_accuracy: 0.8680 - 129ms/epoch - 8ms/step\n", "Epoch 50/400\n", "16/16 - 0s - loss: 0.2494 - accuracy: 0.8980 - val_loss: 0.2866 - val_accuracy: 0.8680 - 68ms/epoch - 4ms/step\n", "Epoch 51/400\n", "16/16 - 0s - loss: 0.2463 - accuracy: 0.9000 - val_loss: 0.2833 - val_accuracy: 0.8700 - 105ms/epoch - 7ms/step\n", "Epoch 52/400\n", "16/16 - 0s - loss: 0.2434 - accuracy: 0.9020 - val_loss: 0.2799 - val_accuracy: 0.8700 - 108ms/epoch - 7ms/step\n", "Epoch 53/400\n", "16/16 - 0s - loss: 0.2404 - accuracy: 0.9040 - val_loss: 0.2766 - val_accuracy: 0.8700 - 72ms/epoch - 4ms/step\n", "Epoch 54/400\n", "16/16 - 0s - loss: 0.2376 - accuracy: 0.9040 - val_loss: 0.2734 - val_accuracy: 0.8720 - 73ms/epoch - 5ms/step\n", "Epoch 55/400\n", "16/16 - 0s - loss: 0.2348 - accuracy: 0.9120 - val_loss: 0.2705 - val_accuracy: 0.8720 - 74ms/epoch - 5ms/step\n", "Epoch 56/400\n", "16/16 - 0s - loss: 0.2322 - accuracy: 0.9140 - val_loss: 0.2675 - val_accuracy: 0.8740 - 71ms/epoch - 4ms/step\n", "Epoch 57/400\n", "16/16 - 0s - loss: 0.2295 - accuracy: 0.9160 - val_loss: 0.2647 - val_accuracy: 0.8780 - 76ms/epoch - 5ms/step\n", "Epoch 58/400\n", "16/16 - 0s - loss: 0.2270 - accuracy: 0.9160 - val_loss: 0.2619 - val_accuracy: 0.8780 - 79ms/epoch - 5ms/step\n", "Epoch 59/400\n", "16/16 - 0s - loss: 0.2246 - accuracy: 0.9160 - val_loss: 0.2590 - val_accuracy: 0.8820 - 77ms/epoch - 5ms/step\n", "Epoch 60/400\n", "16/16 - 0s - loss: 0.2221 - accuracy: 0.9160 - val_loss: 0.2563 - val_accuracy: 0.8860 - 82ms/epoch - 5ms/step\n", "Epoch 61/400\n", "16/16 - 0s - loss: 0.2198 - accuracy: 0.9160 - val_loss: 0.2537 - val_accuracy: 0.8920 - 114ms/epoch - 7ms/step\n", "Epoch 62/400\n", "16/16 - 0s - loss: 0.2176 - accuracy: 0.9180 - val_loss: 0.2512 - val_accuracy: 0.8920 - 82ms/epoch - 5ms/step\n", "Epoch 63/400\n", "16/16 - 0s - loss: 0.2153 - accuracy: 0.9200 - val_loss: 0.2488 - val_accuracy: 0.8920 - 73ms/epoch - 5ms/step\n", "Epoch 64/400\n", "16/16 - 0s - loss: 0.2132 - accuracy: 0.9240 - val_loss: 0.2462 - val_accuracy: 0.8920 - 70ms/epoch - 4ms/step\n", "Epoch 65/400\n", "16/16 - 0s - loss: 0.2111 - accuracy: 0.9240 - val_loss: 0.2439 - val_accuracy: 0.8960 - 76ms/epoch - 5ms/step\n", "Epoch 66/400\n", "16/16 - 0s - loss: 0.2091 - accuracy: 0.9320 - val_loss: 0.2417 - val_accuracy: 0.8980 - 115ms/epoch - 7ms/step\n", "Epoch 67/400\n", "16/16 - 0s - loss: 0.2071 - accuracy: 0.9340 - val_loss: 0.2394 - val_accuracy: 0.9020 - 76ms/epoch - 5ms/step\n", "Epoch 68/400\n", "16/16 - 0s - loss: 0.2051 - accuracy: 0.9360 - val_loss: 0.2372 - val_accuracy: 0.9020 - 81ms/epoch - 5ms/step\n", "Epoch 69/400\n", "16/16 - 0s - loss: 0.2033 - accuracy: 0.9380 - val_loss: 0.2350 - val_accuracy: 0.9020 - 74ms/epoch - 5ms/step\n", "Epoch 70/400\n", "16/16 - 0s - loss: 0.2015 - accuracy: 0.9400 - val_loss: 0.2328 - val_accuracy: 0.9100 - 69ms/epoch - 4ms/step\n", "Epoch 71/400\n", "16/16 - 0s - loss: 0.1996 - accuracy: 0.9440 - val_loss: 0.2309 - val_accuracy: 0.9100 - 70ms/epoch - 4ms/step\n", "Epoch 72/400\n", "16/16 - 0s - loss: 0.1979 - accuracy: 0.9460 - val_loss: 0.2288 - val_accuracy: 0.9140 - 125ms/epoch - 8ms/step\n", "Epoch 73/400\n", "16/16 - 0s - loss: 0.1961 - accuracy: 0.9460 - val_loss: 0.2268 - val_accuracy: 0.9180 - 110ms/epoch - 7ms/step\n", "Epoch 74/400\n", "16/16 - 0s - loss: 0.1945 - accuracy: 0.9460 - val_loss: 0.2248 - val_accuracy: 0.9240 - 69ms/epoch - 4ms/step\n", "Epoch 75/400\n", "16/16 - 0s - loss: 0.1928 - accuracy: 0.9460 - val_loss: 0.2231 - val_accuracy: 0.9240 - 109ms/epoch - 7ms/step\n", "Epoch 76/400\n", "16/16 - 0s - loss: 0.1913 - accuracy: 0.9480 - val_loss: 0.2211 - val_accuracy: 0.9280 - 109ms/epoch - 7ms/step\n", "Epoch 77/400\n", "16/16 - 0s - loss: 0.1897 - accuracy: 0.9480 - val_loss: 0.2193 - val_accuracy: 0.9320 - 74ms/epoch - 5ms/step\n", "Epoch 78/400\n", "16/16 - 0s - loss: 0.1882 - accuracy: 0.9480 - val_loss: 0.2174 - val_accuracy: 0.9320 - 114ms/epoch - 7ms/step\n", "Epoch 79/400\n", "16/16 - 0s - loss: 0.1866 - accuracy: 0.9480 - val_loss: 0.2158 - val_accuracy: 0.9380 - 73ms/epoch - 5ms/step\n", "Epoch 80/400\n", "16/16 - 0s - loss: 0.1852 - accuracy: 0.9480 - val_loss: 0.2140 - val_accuracy: 0.9380 - 110ms/epoch - 7ms/step\n", "Epoch 81/400\n", "16/16 - 0s - loss: 0.1838 - accuracy: 0.9500 - val_loss: 0.2123 - val_accuracy: 0.9380 - 112ms/epoch - 7ms/step\n", "Epoch 82/400\n", "16/16 - 0s - loss: 0.1824 - accuracy: 0.9500 - val_loss: 0.2107 - val_accuracy: 0.9400 - 69ms/epoch - 4ms/step\n", "Epoch 83/400\n", "16/16 - 0s - loss: 0.1810 - accuracy: 0.9540 - val_loss: 0.2090 - val_accuracy: 0.9400 - 69ms/epoch - 4ms/step\n", "Epoch 84/400\n", "16/16 - 0s - loss: 0.1796 - accuracy: 0.9580 - val_loss: 0.2074 - val_accuracy: 0.9440 - 107ms/epoch - 7ms/step\n", "Epoch 85/400\n", "16/16 - 0s - loss: 0.1783 - accuracy: 0.9600 - val_loss: 0.2058 - val_accuracy: 0.9460 - 122ms/epoch - 8ms/step\n", "Epoch 86/400\n", "16/16 - 0s - loss: 0.1770 - accuracy: 0.9600 - val_loss: 0.2043 - val_accuracy: 0.9460 - 76ms/epoch - 5ms/step\n", "Epoch 87/400\n", "16/16 - 0s - loss: 0.1757 - accuracy: 0.9600 - val_loss: 0.2027 - val_accuracy: 0.9460 - 78ms/epoch - 5ms/step\n", "Epoch 88/400\n", "16/16 - 0s - loss: 0.1744 - accuracy: 0.9620 - val_loss: 0.2013 - val_accuracy: 0.9460 - 74ms/epoch - 5ms/step\n", "Epoch 89/400\n", "16/16 - 0s - loss: 0.1732 - accuracy: 0.9640 - val_loss: 0.1997 - val_accuracy: 0.9480 - 73ms/epoch - 5ms/step\n", "Epoch 90/400\n", "16/16 - 0s - loss: 0.1719 - accuracy: 0.9640 - val_loss: 0.1982 - val_accuracy: 0.9480 - 68ms/epoch - 4ms/step\n", "Epoch 91/400\n", "16/16 - 0s - loss: 0.1707 - accuracy: 0.9640 - val_loss: 0.1966 - val_accuracy: 0.9480 - 110ms/epoch - 7ms/step\n", "Epoch 92/400\n", "16/16 - 0s - loss: 0.1695 - accuracy: 0.9640 - val_loss: 0.1953 - val_accuracy: 0.9480 - 76ms/epoch - 5ms/step\n", "Epoch 93/400\n", "16/16 - 0s - loss: 0.1683 - accuracy: 0.9640 - val_loss: 0.1938 - val_accuracy: 0.9520 - 110ms/epoch - 7ms/step\n", "Epoch 94/400\n", "16/16 - 0s - loss: 0.1671 - accuracy: 0.9640 - val_loss: 0.1924 - val_accuracy: 0.9540 - 114ms/epoch - 7ms/step\n", "Epoch 95/400\n", "16/16 - 0s - loss: 0.1659 - accuracy: 0.9640 - val_loss: 0.1911 - val_accuracy: 0.9540 - 67ms/epoch - 4ms/step\n", "Epoch 96/400\n", "16/16 - 0s - loss: 0.1648 - accuracy: 0.9680 - val_loss: 0.1896 - val_accuracy: 0.9540 - 67ms/epoch - 4ms/step\n", "Epoch 97/400\n", "16/16 - 0s - loss: 0.1636 - accuracy: 0.9680 - val_loss: 0.1882 - val_accuracy: 0.9540 - 68ms/epoch - 4ms/step\n", "Epoch 98/400\n", "16/16 - 0s - loss: 0.1625 - accuracy: 0.9700 - val_loss: 0.1868 - val_accuracy: 0.9540 - 71ms/epoch - 4ms/step\n", "Epoch 99/400\n", "16/16 - 0s - loss: 0.1613 - accuracy: 0.9700 - val_loss: 0.1855 - val_accuracy: 0.9560 - 72ms/epoch - 5ms/step\n", "Epoch 100/400\n", "16/16 - 0s - loss: 0.1602 - accuracy: 0.9700 - val_loss: 0.1841 - val_accuracy: 0.9560 - 115ms/epoch - 7ms/step\n", "Epoch 101/400\n", "16/16 - 0s - loss: 0.1591 - accuracy: 0.9700 - val_loss: 0.1827 - val_accuracy: 0.9560 - 70ms/epoch - 4ms/step\n", "Epoch 102/400\n", "16/16 - 0s - loss: 0.1579 - accuracy: 0.9720 - val_loss: 0.1814 - val_accuracy: 0.9560 - 117ms/epoch - 7ms/step\n", "Epoch 103/400\n", "16/16 - 0s - loss: 0.1568 - accuracy: 0.9720 - val_loss: 0.1799 - val_accuracy: 0.9560 - 113ms/epoch - 7ms/step\n", "Epoch 104/400\n", "16/16 - 0s - loss: 0.1557 - accuracy: 0.9720 - val_loss: 0.1785 - val_accuracy: 0.9560 - 124ms/epoch - 8ms/step\n", "Epoch 105/400\n", "16/16 - 0s - loss: 0.1545 - accuracy: 0.9760 - val_loss: 0.1771 - val_accuracy: 0.9580 - 73ms/epoch - 5ms/step\n", "Epoch 106/400\n", "16/16 - 0s - loss: 0.1534 - accuracy: 0.9760 - val_loss: 0.1758 - val_accuracy: 0.9580 - 71ms/epoch - 4ms/step\n", "Epoch 107/400\n", "16/16 - 0s - loss: 0.1523 - accuracy: 0.9760 - val_loss: 0.1744 - val_accuracy: 0.9600 - 116ms/epoch - 7ms/step\n", "Epoch 108/400\n", "16/16 - 0s - loss: 0.1512 - accuracy: 0.9760 - val_loss: 0.1730 - val_accuracy: 0.9600 - 119ms/epoch - 7ms/step\n", "Epoch 109/400\n", "16/16 - 0s - loss: 0.1500 - accuracy: 0.9780 - val_loss: 0.1717 - val_accuracy: 0.9620 - 112ms/epoch - 7ms/step\n", "Epoch 110/400\n", "16/16 - 0s - loss: 0.1489 - accuracy: 0.9780 - val_loss: 0.1703 - val_accuracy: 0.9680 - 69ms/epoch - 4ms/step\n", "Epoch 111/400\n", "16/16 - 0s - loss: 0.1478 - accuracy: 0.9780 - val_loss: 0.1689 - val_accuracy: 0.9680 - 111ms/epoch - 7ms/step\n", "Epoch 112/400\n", "16/16 - 0s - loss: 0.1467 - accuracy: 0.9780 - val_loss: 0.1676 - val_accuracy: 0.9680 - 116ms/epoch - 7ms/step\n", "Epoch 113/400\n", "16/16 - 0s - loss: 0.1455 - accuracy: 0.9800 - val_loss: 0.1662 - val_accuracy: 0.9720 - 110ms/epoch - 7ms/step\n", "Epoch 114/400\n", "16/16 - 0s - loss: 0.1444 - accuracy: 0.9800 - val_loss: 0.1649 - val_accuracy: 0.9720 - 65ms/epoch - 4ms/step\n", "Epoch 115/400\n", "16/16 - 0s - loss: 0.1433 - accuracy: 0.9800 - val_loss: 0.1635 - val_accuracy: 0.9720 - 69ms/epoch - 4ms/step\n", "Epoch 116/400\n", "16/16 - 0s - loss: 0.1422 - accuracy: 0.9800 - val_loss: 0.1620 - val_accuracy: 0.9720 - 71ms/epoch - 4ms/step\n", "Epoch 117/400\n", "16/16 - 0s - loss: 0.1410 - accuracy: 0.9800 - val_loss: 0.1607 - val_accuracy: 0.9740 - 75ms/epoch - 5ms/step\n", "Epoch 118/400\n", "16/16 - 0s - loss: 0.1399 - accuracy: 0.9800 - val_loss: 0.1594 - val_accuracy: 0.9760 - 109ms/epoch - 7ms/step\n", "Epoch 119/400\n", "16/16 - 0s - loss: 0.1388 - accuracy: 0.9800 - val_loss: 0.1580 - val_accuracy: 0.9760 - 87ms/epoch - 5ms/step\n", "Epoch 120/400\n", "16/16 - 0s - loss: 0.1376 - accuracy: 0.9800 - val_loss: 0.1567 - val_accuracy: 0.9760 - 115ms/epoch - 7ms/step\n", "Epoch 121/400\n", "16/16 - 0s - loss: 0.1365 - accuracy: 0.9800 - val_loss: 0.1554 - val_accuracy: 0.9760 - 74ms/epoch - 5ms/step\n", "Epoch 122/400\n", "16/16 - 0s - loss: 0.1354 - accuracy: 0.9800 - val_loss: 0.1541 - val_accuracy: 0.9760 - 73ms/epoch - 5ms/step\n", "Epoch 123/400\n", "16/16 - 0s - loss: 0.1343 - accuracy: 0.9820 - val_loss: 0.1527 - val_accuracy: 0.9760 - 69ms/epoch - 4ms/step\n", "Epoch 124/400\n", "16/16 - 0s - loss: 0.1331 - accuracy: 0.9820 - val_loss: 0.1514 - val_accuracy: 0.9760 - 75ms/epoch - 5ms/step\n", "Epoch 125/400\n", "16/16 - 0s - loss: 0.1320 - accuracy: 0.9820 - val_loss: 0.1500 - val_accuracy: 0.9760 - 123ms/epoch - 8ms/step\n", "Epoch 126/400\n", "16/16 - 0s - loss: 0.1309 - accuracy: 0.9820 - val_loss: 0.1486 - val_accuracy: 0.9760 - 111ms/epoch - 7ms/step\n", "Epoch 127/400\n", "16/16 - 0s - loss: 0.1297 - accuracy: 0.9820 - val_loss: 0.1473 - val_accuracy: 0.9760 - 70ms/epoch - 4ms/step\n", "Epoch 128/400\n", "16/16 - 0s - loss: 0.1287 - accuracy: 0.9820 - val_loss: 0.1459 - val_accuracy: 0.9760 - 69ms/epoch - 4ms/step\n", "Epoch 129/400\n", "16/16 - 0s - loss: 0.1275 - accuracy: 0.9820 - val_loss: 0.1446 - val_accuracy: 0.9760 - 118ms/epoch - 7ms/step\n", "Epoch 130/400\n", "16/16 - 0s - loss: 0.1264 - accuracy: 0.9820 - val_loss: 0.1433 - val_accuracy: 0.9760 - 74ms/epoch - 5ms/step\n", "Epoch 131/400\n", "16/16 - 0s - loss: 0.1253 - accuracy: 0.9820 - val_loss: 0.1420 - val_accuracy: 0.9780 - 120ms/epoch - 8ms/step\n", "Epoch 132/400\n", "16/16 - 0s - loss: 0.1242 - accuracy: 0.9820 - val_loss: 0.1407 - val_accuracy: 0.9780 - 74ms/epoch - 5ms/step\n", "Epoch 133/400\n", "16/16 - 0s - loss: 0.1231 - accuracy: 0.9840 - val_loss: 0.1393 - val_accuracy: 0.9780 - 74ms/epoch - 5ms/step\n", "Epoch 134/400\n", "16/16 - 0s - loss: 0.1219 - accuracy: 0.9840 - val_loss: 0.1381 - val_accuracy: 0.9780 - 73ms/epoch - 5ms/step\n", "Epoch 135/400\n", "16/16 - 0s - loss: 0.1209 - accuracy: 0.9840 - val_loss: 0.1367 - val_accuracy: 0.9800 - 79ms/epoch - 5ms/step\n", "Epoch 136/400\n", "16/16 - 0s - loss: 0.1198 - accuracy: 0.9860 - val_loss: 0.1355 - val_accuracy: 0.9800 - 81ms/epoch - 5ms/step\n", "Epoch 137/400\n", "16/16 - 0s - loss: 0.1187 - accuracy: 0.9860 - val_loss: 0.1343 - val_accuracy: 0.9800 - 71ms/epoch - 4ms/step\n", "Epoch 138/400\n", "16/16 - 0s - loss: 0.1176 - accuracy: 0.9860 - val_loss: 0.1330 - val_accuracy: 0.9820 - 74ms/epoch - 5ms/step\n", "Epoch 139/400\n", "16/16 - 0s - loss: 0.1166 - accuracy: 0.9860 - val_loss: 0.1318 - val_accuracy: 0.9820 - 111ms/epoch - 7ms/step\n", "Epoch 140/400\n", "16/16 - 0s - loss: 0.1155 - accuracy: 0.9860 - val_loss: 0.1307 - val_accuracy: 0.9820 - 70ms/epoch - 4ms/step\n", "Epoch 141/400\n", "16/16 - 0s - loss: 0.1144 - accuracy: 0.9860 - val_loss: 0.1295 - val_accuracy: 0.9840 - 73ms/epoch - 5ms/step\n", "Epoch 142/400\n", "16/16 - 0s - loss: 0.1134 - accuracy: 0.9860 - val_loss: 0.1282 - val_accuracy: 0.9840 - 114ms/epoch - 7ms/step\n", "Epoch 143/400\n", "16/16 - 0s - loss: 0.1124 - accuracy: 0.9880 - val_loss: 0.1271 - val_accuracy: 0.9840 - 70ms/epoch - 4ms/step\n", "Epoch 144/400\n", "16/16 - 0s - loss: 0.1114 - accuracy: 0.9880 - val_loss: 0.1259 - val_accuracy: 0.9840 - 72ms/epoch - 5ms/step\n", "Epoch 145/400\n", "16/16 - 0s - loss: 0.1103 - accuracy: 0.9880 - val_loss: 0.1247 - val_accuracy: 0.9840 - 74ms/epoch - 5ms/step\n", "Epoch 146/400\n", "16/16 - 0s - loss: 0.1093 - accuracy: 0.9880 - val_loss: 0.1236 - val_accuracy: 0.9860 - 71ms/epoch - 4ms/step\n", "Epoch 147/400\n", "16/16 - 0s - loss: 0.1083 - accuracy: 0.9880 - val_loss: 0.1224 - val_accuracy: 0.9860 - 80ms/epoch - 5ms/step\n", "Epoch 148/400\n", "16/16 - 0s - loss: 0.1073 - accuracy: 0.9880 - val_loss: 0.1212 - val_accuracy: 0.9860 - 79ms/epoch - 5ms/step\n", "Epoch 149/400\n", "16/16 - 0s - loss: 0.1063 - accuracy: 0.9880 - val_loss: 0.1202 - val_accuracy: 0.9860 - 72ms/epoch - 4ms/step\n", "Epoch 150/400\n", "16/16 - 0s - loss: 0.1054 - accuracy: 0.9860 - val_loss: 0.1191 - val_accuracy: 0.9860 - 70ms/epoch - 4ms/step\n", "Epoch 151/400\n", "16/16 - 0s - loss: 0.1044 - accuracy: 0.9860 - val_loss: 0.1180 - val_accuracy: 0.9860 - 114ms/epoch - 7ms/step\n", "Epoch 152/400\n", "16/16 - 0s - loss: 0.1034 - accuracy: 0.9860 - val_loss: 0.1168 - val_accuracy: 0.9860 - 73ms/epoch - 5ms/step\n", "Epoch 153/400\n", "16/16 - 0s - loss: 0.1025 - accuracy: 0.9860 - val_loss: 0.1158 - val_accuracy: 0.9860 - 76ms/epoch - 5ms/step\n", "Epoch 154/400\n", "16/16 - 0s - loss: 0.1016 - accuracy: 0.9860 - val_loss: 0.1147 - val_accuracy: 0.9860 - 75ms/epoch - 5ms/step\n", "Epoch 155/400\n", "16/16 - 0s - loss: 0.1006 - accuracy: 0.9860 - val_loss: 0.1136 - val_accuracy: 0.9860 - 74ms/epoch - 5ms/step\n", "Epoch 156/400\n", "16/16 - 0s - loss: 0.0997 - accuracy: 0.9860 - val_loss: 0.1126 - val_accuracy: 0.9860 - 115ms/epoch - 7ms/step\n", "Epoch 157/400\n", "16/16 - 0s - loss: 0.0988 - accuracy: 0.9860 - val_loss: 0.1116 - val_accuracy: 0.9860 - 73ms/epoch - 5ms/step\n", "Epoch 158/400\n", "16/16 - 0s - loss: 0.0979 - accuracy: 0.9860 - val_loss: 0.1106 - val_accuracy: 0.9860 - 223ms/epoch - 14ms/step\n", "Epoch 159/400\n", "16/16 - 0s - loss: 0.0970 - accuracy: 0.9860 - val_loss: 0.1097 - val_accuracy: 0.9860 - 287ms/epoch - 18ms/step\n", "Epoch 160/400\n", "16/16 - 0s - loss: 0.0962 - accuracy: 0.9860 - val_loss: 0.1086 - val_accuracy: 0.9860 - 209ms/epoch - 13ms/step\n", "Epoch 161/400\n", "16/16 - 0s - loss: 0.0953 - accuracy: 0.9860 - val_loss: 0.1076 - val_accuracy: 0.9880 - 80ms/epoch - 5ms/step\n", "Epoch 162/400\n", "16/16 - 0s - loss: 0.0944 - accuracy: 0.9860 - val_loss: 0.1067 - val_accuracy: 0.9900 - 109ms/epoch - 7ms/step\n", "Epoch 163/400\n", "16/16 - 0s - loss: 0.0936 - accuracy: 0.9860 - val_loss: 0.1057 - val_accuracy: 0.9900 - 70ms/epoch - 4ms/step\n", "Epoch 164/400\n", "16/16 - 0s - loss: 0.0927 - accuracy: 0.9860 - val_loss: 0.1048 - val_accuracy: 0.9920 - 71ms/epoch - 4ms/step\n", "Epoch 165/400\n", "16/16 - 0s - loss: 0.0919 - accuracy: 0.9860 - val_loss: 0.1039 - val_accuracy: 0.9920 - 77ms/epoch - 5ms/step\n", "Epoch 166/400\n", "16/16 - 0s - loss: 0.0911 - accuracy: 0.9860 - val_loss: 0.1030 - val_accuracy: 0.9920 - 226ms/epoch - 14ms/step\n", "Epoch 167/400\n", "16/16 - 0s - loss: 0.0903 - accuracy: 0.9860 - val_loss: 0.1021 - val_accuracy: 0.9920 - 332ms/epoch - 21ms/step\n", "Epoch 168/400\n", "16/16 - 0s - loss: 0.0894 - accuracy: 0.9880 - val_loss: 0.1012 - val_accuracy: 0.9920 - 179ms/epoch - 11ms/step\n", "Epoch 169/400\n", "16/16 - 0s - loss: 0.0887 - accuracy: 0.9900 - val_loss: 0.1003 - val_accuracy: 0.9920 - 69ms/epoch - 4ms/step\n", "Epoch 170/400\n", "16/16 - 0s - loss: 0.0879 - accuracy: 0.9900 - val_loss: 0.0995 - val_accuracy: 0.9920 - 76ms/epoch - 5ms/step\n", "Epoch 171/400\n", "16/16 - 0s - loss: 0.0871 - accuracy: 0.9900 - val_loss: 0.0986 - val_accuracy: 0.9920 - 68ms/epoch - 4ms/step\n", "Epoch 172/400\n", "16/16 - 0s - loss: 0.0863 - accuracy: 0.9920 - val_loss: 0.0978 - val_accuracy: 0.9920 - 70ms/epoch - 4ms/step\n", "Epoch 173/400\n", "16/16 - 0s - loss: 0.0856 - accuracy: 0.9920 - val_loss: 0.0969 - val_accuracy: 0.9940 - 76ms/epoch - 5ms/step\n", "Epoch 174/400\n", "16/16 - 0s - loss: 0.0849 - accuracy: 0.9920 - val_loss: 0.0961 - val_accuracy: 0.9940 - 352ms/epoch - 22ms/step\n", "Epoch 175/400\n", "16/16 - 0s - loss: 0.0841 - accuracy: 0.9920 - val_loss: 0.0953 - val_accuracy: 0.9940 - 228ms/epoch - 14ms/step\n", "Epoch 176/400\n", "16/16 - 0s - loss: 0.0834 - accuracy: 0.9920 - val_loss: 0.0944 - val_accuracy: 0.9940 - 98ms/epoch - 6ms/step\n", "Epoch 177/400\n", "16/16 - 0s - loss: 0.0826 - accuracy: 0.9920 - val_loss: 0.0937 - val_accuracy: 0.9940 - 69ms/epoch - 4ms/step\n", "Epoch 178/400\n", "16/16 - 0s - loss: 0.0819 - accuracy: 0.9920 - val_loss: 0.0929 - val_accuracy: 0.9940 - 71ms/epoch - 4ms/step\n", "Epoch 179/400\n", "16/16 - 0s - loss: 0.0812 - accuracy: 0.9920 - val_loss: 0.0920 - val_accuracy: 0.9940 - 73ms/epoch - 5ms/step\n", "Epoch 180/400\n", "16/16 - 0s - loss: 0.0805 - accuracy: 0.9920 - val_loss: 0.0913 - val_accuracy: 0.9940 - 124ms/epoch - 8ms/step\n", "Epoch 181/400\n", "16/16 - 0s - loss: 0.0799 - accuracy: 0.9920 - val_loss: 0.0905 - val_accuracy: 0.9940 - 115ms/epoch - 7ms/step\n", "Epoch 182/400\n", "16/16 - 0s - loss: 0.0792 - accuracy: 0.9920 - val_loss: 0.0899 - val_accuracy: 0.9940 - 68ms/epoch - 4ms/step\n", "Epoch 183/400\n", "16/16 - 0s - loss: 0.0785 - accuracy: 0.9920 - val_loss: 0.0891 - val_accuracy: 0.9940 - 112ms/epoch - 7ms/step\n", "Epoch 184/400\n", "16/16 - 0s - loss: 0.0779 - accuracy: 0.9920 - val_loss: 0.0884 - val_accuracy: 0.9940 - 113ms/epoch - 7ms/step\n", "Epoch 185/400\n", "16/16 - 0s - loss: 0.0772 - accuracy: 0.9900 - val_loss: 0.0877 - val_accuracy: 0.9940 - 71ms/epoch - 4ms/step\n", "Epoch 186/400\n", "16/16 - 0s - loss: 0.0766 - accuracy: 0.9900 - val_loss: 0.0870 - val_accuracy: 0.9940 - 75ms/epoch - 5ms/step\n", "Epoch 187/400\n", "16/16 - 0s - loss: 0.0760 - accuracy: 0.9900 - val_loss: 0.0863 - val_accuracy: 0.9920 - 69ms/epoch - 4ms/step\n", "Epoch 188/400\n", "16/16 - 0s - loss: 0.0754 - accuracy: 0.9900 - val_loss: 0.0856 - val_accuracy: 0.9920 - 69ms/epoch - 4ms/step\n", "Epoch 189/400\n", "16/16 - 0s - loss: 0.0748 - accuracy: 0.9900 - val_loss: 0.0849 - val_accuracy: 0.9920 - 71ms/epoch - 4ms/step\n", "Epoch 190/400\n", "16/16 - 0s - loss: 0.0742 - accuracy: 0.9900 - val_loss: 0.0843 - val_accuracy: 0.9920 - 73ms/epoch - 5ms/step\n", "Epoch 191/400\n", "16/16 - 0s - loss: 0.0736 - accuracy: 0.9900 - val_loss: 0.0837 - val_accuracy: 0.9920 - 119ms/epoch - 7ms/step\n", "Epoch 192/400\n", "16/16 - 0s - loss: 0.0730 - accuracy: 0.9900 - val_loss: 0.0830 - val_accuracy: 0.9920 - 69ms/epoch - 4ms/step\n", "Epoch 193/400\n", "16/16 - 0s - loss: 0.0724 - accuracy: 0.9940 - val_loss: 0.0824 - val_accuracy: 0.9920 - 76ms/epoch - 5ms/step\n", "Epoch 194/400\n", "16/16 - 0s - loss: 0.0718 - accuracy: 0.9940 - val_loss: 0.0818 - val_accuracy: 0.9920 - 69ms/epoch - 4ms/step\n", "Epoch 195/400\n", "16/16 - 0s - loss: 0.0713 - accuracy: 0.9940 - val_loss: 0.0811 - val_accuracy: 0.9940 - 71ms/epoch - 4ms/step\n", "Epoch 196/400\n", "16/16 - 0s - loss: 0.0707 - accuracy: 0.9940 - val_loss: 0.0805 - val_accuracy: 0.9940 - 70ms/epoch - 4ms/step\n", "Epoch 197/400\n", "16/16 - 0s - loss: 0.0702 - accuracy: 0.9940 - val_loss: 0.0799 - val_accuracy: 0.9940 - 72ms/epoch - 4ms/step\n", "Epoch 198/400\n", "16/16 - 0s - loss: 0.0696 - accuracy: 0.9940 - val_loss: 0.0793 - val_accuracy: 0.9940 - 73ms/epoch - 5ms/step\n", "Epoch 199/400\n", "16/16 - 0s - loss: 0.0691 - accuracy: 0.9940 - val_loss: 0.0787 - val_accuracy: 0.9940 - 109ms/epoch - 7ms/step\n", "Epoch 200/400\n", "16/16 - 0s - loss: 0.0686 - accuracy: 0.9940 - val_loss: 0.0782 - val_accuracy: 0.9960 - 110ms/epoch - 7ms/step\n", "Epoch 201/400\n", "16/16 - 0s - loss: 0.0681 - accuracy: 0.9940 - val_loss: 0.0776 - val_accuracy: 0.9960 - 77ms/epoch - 5ms/step\n", "Epoch 202/400\n", "16/16 - 0s - loss: 0.0676 - accuracy: 0.9940 - val_loss: 0.0770 - val_accuracy: 0.9960 - 66ms/epoch - 4ms/step\n", "Epoch 203/400\n", "16/16 - 0s - loss: 0.0671 - accuracy: 0.9940 - val_loss: 0.0765 - val_accuracy: 0.9960 - 77ms/epoch - 5ms/step\n", "Epoch 204/400\n", "16/16 - 0s - loss: 0.0666 - accuracy: 0.9940 - val_loss: 0.0759 - val_accuracy: 0.9960 - 69ms/epoch - 4ms/step\n", "Epoch 205/400\n", "16/16 - 0s - loss: 0.0661 - accuracy: 0.9940 - val_loss: 0.0753 - val_accuracy: 0.9960 - 74ms/epoch - 5ms/step\n", "Epoch 206/400\n", "16/16 - 0s - loss: 0.0656 - accuracy: 0.9940 - val_loss: 0.0749 - val_accuracy: 0.9960 - 72ms/epoch - 5ms/step\n", "Epoch 207/400\n", "16/16 - 0s - loss: 0.0652 - accuracy: 0.9940 - val_loss: 0.0743 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 208/400\n", "16/16 - 0s - loss: 0.0647 - accuracy: 0.9940 - val_loss: 0.0738 - val_accuracy: 0.9960 - 74ms/epoch - 5ms/step\n", "Epoch 209/400\n", "16/16 - 0s - loss: 0.0642 - accuracy: 0.9940 - val_loss: 0.0733 - val_accuracy: 0.9960 - 67ms/epoch - 4ms/step\n", "Epoch 210/400\n", "16/16 - 0s - loss: 0.0638 - accuracy: 0.9940 - val_loss: 0.0728 - val_accuracy: 0.9960 - 75ms/epoch - 5ms/step\n", "Epoch 211/400\n", "16/16 - 0s - loss: 0.0633 - accuracy: 0.9940 - val_loss: 0.0723 - val_accuracy: 0.9960 - 114ms/epoch - 7ms/step\n", "Epoch 212/400\n", "16/16 - 0s - loss: 0.0629 - accuracy: 0.9940 - val_loss: 0.0718 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 213/400\n", "16/16 - 0s - loss: 0.0624 - accuracy: 0.9940 - val_loss: 0.0713 - val_accuracy: 0.9960 - 128ms/epoch - 8ms/step\n", "Epoch 214/400\n", "16/16 - 0s - loss: 0.0620 - accuracy: 0.9940 - val_loss: 0.0708 - val_accuracy: 0.9960 - 88ms/epoch - 6ms/step\n", "Epoch 215/400\n", "16/16 - 0s - loss: 0.0616 - accuracy: 0.9940 - val_loss: 0.0703 - val_accuracy: 0.9960 - 79ms/epoch - 5ms/step\n", "Epoch 216/400\n", "16/16 - 0s - loss: 0.0611 - accuracy: 0.9940 - val_loss: 0.0698 - val_accuracy: 0.9960 - 114ms/epoch - 7ms/step\n", "Epoch 217/400\n", "16/16 - 0s - loss: 0.0607 - accuracy: 0.9940 - val_loss: 0.0694 - val_accuracy: 0.9960 - 76ms/epoch - 5ms/step\n", "Epoch 218/400\n", "16/16 - 0s - loss: 0.0603 - accuracy: 0.9940 - val_loss: 0.0689 - val_accuracy: 0.9960 - 114ms/epoch - 7ms/step\n", "Epoch 219/400\n", "16/16 - 0s - loss: 0.0599 - accuracy: 0.9940 - val_loss: 0.0685 - val_accuracy: 0.9960 - 118ms/epoch - 7ms/step\n", "Epoch 220/400\n", "16/16 - 0s - loss: 0.0595 - accuracy: 0.9940 - val_loss: 0.0680 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 221/400\n", "16/16 - 0s - loss: 0.0591 - accuracy: 0.9940 - val_loss: 0.0676 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 222/400\n", "16/16 - 0s - loss: 0.0587 - accuracy: 0.9940 - val_loss: 0.0672 - val_accuracy: 0.9960 - 113ms/epoch - 7ms/step\n", "Epoch 223/400\n", "16/16 - 0s - loss: 0.0583 - accuracy: 0.9940 - val_loss: 0.0668 - val_accuracy: 0.9960 - 72ms/epoch - 4ms/step\n", "Epoch 224/400\n", "16/16 - 0s - loss: 0.0580 - accuracy: 0.9940 - val_loss: 0.0663 - val_accuracy: 0.9960 - 74ms/epoch - 5ms/step\n", "Epoch 225/400\n", "16/16 - 0s - loss: 0.0576 - accuracy: 0.9940 - val_loss: 0.0659 - val_accuracy: 0.9960 - 124ms/epoch - 8ms/step\n", "Epoch 226/400\n", "16/16 - 0s - loss: 0.0572 - accuracy: 0.9940 - val_loss: 0.0655 - val_accuracy: 0.9960 - 114ms/epoch - 7ms/step\n", "Epoch 227/400\n", "16/16 - 0s - loss: 0.0568 - accuracy: 0.9940 - val_loss: 0.0650 - val_accuracy: 0.9960 - 69ms/epoch - 4ms/step\n", "Epoch 228/400\n", "16/16 - 0s - loss: 0.0565 - accuracy: 0.9940 - val_loss: 0.0647 - val_accuracy: 0.9960 - 110ms/epoch - 7ms/step\n", "Epoch 229/400\n", "16/16 - 0s - loss: 0.0561 - accuracy: 0.9940 - val_loss: 0.0642 - val_accuracy: 0.9960 - 112ms/epoch - 7ms/step\n", "Epoch 230/400\n", "16/16 - 0s - loss: 0.0558 - accuracy: 0.9940 - val_loss: 0.0639 - val_accuracy: 0.9960 - 77ms/epoch - 5ms/step\n", "Epoch 231/400\n", "16/16 - 0s - loss: 0.0554 - accuracy: 0.9940 - val_loss: 0.0635 - val_accuracy: 0.9940 - 68ms/epoch - 4ms/step\n", "Epoch 232/400\n", "16/16 - 0s - loss: 0.0550 - accuracy: 0.9940 - val_loss: 0.0630 - val_accuracy: 0.9960 - 70ms/epoch - 4ms/step\n", "Epoch 233/400\n", "16/16 - 0s - loss: 0.0547 - accuracy: 0.9940 - val_loss: 0.0626 - val_accuracy: 0.9940 - 70ms/epoch - 4ms/step\n", "Epoch 234/400\n", "16/16 - 0s - loss: 0.0543 - accuracy: 0.9940 - val_loss: 0.0623 - val_accuracy: 0.9940 - 111ms/epoch - 7ms/step\n", "Epoch 235/400\n", "16/16 - 0s - loss: 0.0540 - accuracy: 0.9940 - val_loss: 0.0619 - val_accuracy: 0.9940 - 118ms/epoch - 7ms/step\n", "Epoch 236/400\n", "16/16 - 0s - loss: 0.0537 - accuracy: 0.9940 - val_loss: 0.0615 - val_accuracy: 0.9940 - 80ms/epoch - 5ms/step\n", "Epoch 237/400\n", "16/16 - 0s - loss: 0.0534 - accuracy: 0.9940 - val_loss: 0.0611 - val_accuracy: 0.9940 - 108ms/epoch - 7ms/step\n", "Epoch 238/400\n", "16/16 - 0s - loss: 0.0530 - accuracy: 0.9940 - val_loss: 0.0608 - val_accuracy: 0.9940 - 72ms/epoch - 5ms/step\n", "Epoch 239/400\n", "16/16 - 0s - loss: 0.0527 - accuracy: 0.9940 - val_loss: 0.0604 - val_accuracy: 0.9940 - 112ms/epoch - 7ms/step\n", "Epoch 240/400\n", "16/16 - 0s - loss: 0.0524 - accuracy: 0.9940 - val_loss: 0.0600 - val_accuracy: 0.9940 - 109ms/epoch - 7ms/step\n", "Epoch 241/400\n", "16/16 - 0s - loss: 0.0521 - accuracy: 0.9940 - val_loss: 0.0597 - val_accuracy: 0.9940 - 68ms/epoch - 4ms/step\n", "Epoch 242/400\n", "16/16 - 0s - loss: 0.0518 - accuracy: 0.9940 - val_loss: 0.0593 - val_accuracy: 0.9940 - 71ms/epoch - 4ms/step\n", "Epoch 243/400\n", "16/16 - 0s - loss: 0.0515 - accuracy: 0.9940 - val_loss: 0.0590 - val_accuracy: 0.9940 - 68ms/epoch - 4ms/step\n", "Epoch 244/400\n", "16/16 - 0s - loss: 0.0511 - accuracy: 0.9940 - val_loss: 0.0587 - val_accuracy: 0.9940 - 79ms/epoch - 5ms/step\n", "Epoch 245/400\n", "16/16 - 0s - loss: 0.0508 - accuracy: 0.9940 - val_loss: 0.0583 - val_accuracy: 0.9940 - 77ms/epoch - 5ms/step\n", "Epoch 246/400\n", "16/16 - 0s - loss: 0.0505 - accuracy: 0.9940 - val_loss: 0.0579 - val_accuracy: 0.9940 - 113ms/epoch - 7ms/step\n", "Epoch 247/400\n", "16/16 - 0s - loss: 0.0502 - accuracy: 0.9940 - val_loss: 0.0576 - val_accuracy: 0.9940 - 113ms/epoch - 7ms/step\n", "Epoch 248/400\n", "16/16 - 0s - loss: 0.0500 - accuracy: 0.9940 - val_loss: 0.0573 - val_accuracy: 0.9940 - 109ms/epoch - 7ms/step\n", "Epoch 249/400\n", "16/16 - 0s - loss: 0.0496 - accuracy: 0.9940 - val_loss: 0.0570 - val_accuracy: 0.9940 - 68ms/epoch - 4ms/step\n", "Epoch 250/400\n", "16/16 - 0s - loss: 0.0494 - accuracy: 0.9940 - val_loss: 0.0567 - val_accuracy: 0.9940 - 74ms/epoch - 5ms/step\n", "Epoch 251/400\n", "16/16 - 0s - loss: 0.0491 - accuracy: 0.9940 - val_loss: 0.0563 - val_accuracy: 0.9940 - 70ms/epoch - 4ms/step\n", "Epoch 252/400\n", "16/16 - 0s - loss: 0.0488 - accuracy: 0.9940 - val_loss: 0.0560 - val_accuracy: 0.9960 - 116ms/epoch - 7ms/step\n", "Epoch 253/400\n", "16/16 - 0s - loss: 0.0485 - accuracy: 0.9940 - val_loss: 0.0556 - val_accuracy: 0.9960 - 72ms/epoch - 4ms/step\n", "Epoch 254/400\n", "16/16 - 0s - loss: 0.0482 - accuracy: 0.9940 - val_loss: 0.0553 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 255/400\n", "16/16 - 0s - loss: 0.0480 - accuracy: 0.9940 - val_loss: 0.0550 - val_accuracy: 0.9960 - 69ms/epoch - 4ms/step\n", "Epoch 256/400\n", "16/16 - 0s - loss: 0.0477 - accuracy: 0.9940 - val_loss: 0.0547 - val_accuracy: 0.9960 - 116ms/epoch - 7ms/step\n", "Epoch 257/400\n", "16/16 - 0s - loss: 0.0474 - accuracy: 0.9940 - val_loss: 0.0544 - val_accuracy: 0.9960 - 82ms/epoch - 5ms/step\n", "Epoch 258/400\n", "16/16 - 0s - loss: 0.0472 - accuracy: 0.9940 - val_loss: 0.0541 - val_accuracy: 0.9960 - 76ms/epoch - 5ms/step\n", "Epoch 259/400\n", "16/16 - 0s - loss: 0.0469 - accuracy: 0.9940 - val_loss: 0.0538 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 260/400\n", "16/16 - 0s - loss: 0.0466 - accuracy: 0.9940 - val_loss: 0.0534 - val_accuracy: 0.9960 - 71ms/epoch - 4ms/step\n", "Epoch 261/400\n", "16/16 - 0s - loss: 0.0463 - accuracy: 0.9940 - val_loss: 0.0532 - val_accuracy: 0.9960 - 69ms/epoch - 4ms/step\n", "Epoch 262/400\n", "16/16 - 0s - loss: 0.0461 - accuracy: 0.9940 - val_loss: 0.0529 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 263/400\n", "16/16 - 0s - loss: 0.0458 - accuracy: 0.9940 - val_loss: 0.0526 - val_accuracy: 0.9960 - 72ms/epoch - 4ms/step\n", "Epoch 264/400\n", "16/16 - 0s - loss: 0.0456 - accuracy: 0.9940 - val_loss: 0.0524 - val_accuracy: 0.9960 - 75ms/epoch - 5ms/step\n", "Epoch 265/400\n", "16/16 - 0s - loss: 0.0453 - accuracy: 0.9940 - val_loss: 0.0520 - val_accuracy: 0.9960 - 114ms/epoch - 7ms/step\n", "Epoch 266/400\n", "16/16 - 0s - loss: 0.0450 - accuracy: 0.9940 - val_loss: 0.0518 - val_accuracy: 0.9960 - 67ms/epoch - 4ms/step\n", "Epoch 267/400\n", "16/16 - 0s - loss: 0.0448 - accuracy: 0.9940 - val_loss: 0.0515 - val_accuracy: 0.9960 - 70ms/epoch - 4ms/step\n", "Epoch 268/400\n", "16/16 - 0s - loss: 0.0445 - accuracy: 0.9940 - val_loss: 0.0512 - val_accuracy: 0.9960 - 74ms/epoch - 5ms/step\n", "Epoch 269/400\n", "16/16 - 0s - loss: 0.0443 - accuracy: 0.9940 - val_loss: 0.0510 - val_accuracy: 0.9960 - 115ms/epoch - 7ms/step\n", "Epoch 270/400\n", "16/16 - 0s - loss: 0.0441 - accuracy: 0.9940 - val_loss: 0.0506 - val_accuracy: 0.9960 - 115ms/epoch - 7ms/step\n", "Epoch 271/400\n", "16/16 - 0s - loss: 0.0438 - accuracy: 0.9940 - val_loss: 0.0503 - val_accuracy: 0.9960 - 70ms/epoch - 4ms/step\n", "Epoch 272/400\n", "16/16 - 0s - loss: 0.0436 - accuracy: 0.9940 - val_loss: 0.0501 - val_accuracy: 0.9960 - 74ms/epoch - 5ms/step\n", "Epoch 273/400\n", "16/16 - 0s - loss: 0.0433 - accuracy: 0.9940 - val_loss: 0.0498 - val_accuracy: 0.9960 - 76ms/epoch - 5ms/step\n", "Epoch 274/400\n", "16/16 - 0s - loss: 0.0431 - accuracy: 0.9940 - val_loss: 0.0495 - val_accuracy: 0.9960 - 121ms/epoch - 8ms/step\n", "Epoch 275/400\n", "16/16 - 0s - loss: 0.0429 - accuracy: 0.9940 - val_loss: 0.0493 - val_accuracy: 0.9960 - 70ms/epoch - 4ms/step\n", "Epoch 276/400\n", "16/16 - 0s - loss: 0.0426 - accuracy: 0.9940 - val_loss: 0.0490 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 277/400\n", "16/16 - 0s - loss: 0.0424 - accuracy: 0.9940 - val_loss: 0.0487 - val_accuracy: 0.9960 - 70ms/epoch - 4ms/step\n", "Epoch 278/400\n", "16/16 - 0s - loss: 0.0422 - accuracy: 0.9940 - val_loss: 0.0484 - val_accuracy: 0.9960 - 116ms/epoch - 7ms/step\n", "Epoch 279/400\n", "16/16 - 0s - loss: 0.0419 - accuracy: 0.9940 - val_loss: 0.0481 - val_accuracy: 0.9960 - 71ms/epoch - 4ms/step\n", "Epoch 280/400\n", "16/16 - 0s - loss: 0.0417 - accuracy: 0.9940 - val_loss: 0.0479 - val_accuracy: 0.9960 - 74ms/epoch - 5ms/step\n", "Epoch 281/400\n", "16/16 - 0s - loss: 0.0415 - accuracy: 0.9940 - val_loss: 0.0476 - val_accuracy: 0.9960 - 80ms/epoch - 5ms/step\n", "Epoch 282/400\n", "16/16 - 0s - loss: 0.0413 - accuracy: 0.9940 - val_loss: 0.0473 - val_accuracy: 0.9960 - 108ms/epoch - 7ms/step\n", "Epoch 283/400\n", "16/16 - 0s - loss: 0.0410 - accuracy: 0.9940 - val_loss: 0.0471 - val_accuracy: 0.9960 - 75ms/epoch - 5ms/step\n", "Epoch 284/400\n", "16/16 - 0s - loss: 0.0408 - accuracy: 0.9940 - val_loss: 0.0469 - val_accuracy: 0.9960 - 115ms/epoch - 7ms/step\n", "Epoch 285/400\n", "16/16 - 0s - loss: 0.0406 - accuracy: 0.9940 - val_loss: 0.0466 - val_accuracy: 0.9960 - 114ms/epoch - 7ms/step\n", "Epoch 286/400\n", "16/16 - 0s - loss: 0.0404 - accuracy: 0.9940 - val_loss: 0.0464 - val_accuracy: 0.9960 - 69ms/epoch - 4ms/step\n", "Epoch 287/400\n", "16/16 - 0s - loss: 0.0402 - accuracy: 0.9940 - val_loss: 0.0461 - val_accuracy: 0.9960 - 71ms/epoch - 4ms/step\n", "Epoch 288/400\n", "16/16 - 0s - loss: 0.0399 - accuracy: 0.9940 - val_loss: 0.0459 - val_accuracy: 0.9960 - 118ms/epoch - 7ms/step\n", "Epoch 289/400\n", "16/16 - 0s - loss: 0.0397 - accuracy: 0.9940 - val_loss: 0.0456 - val_accuracy: 0.9960 - 114ms/epoch - 7ms/step\n", "Epoch 290/400\n", "16/16 - 0s - loss: 0.0395 - accuracy: 0.9940 - val_loss: 0.0454 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 291/400\n", "16/16 - 0s - loss: 0.0393 - accuracy: 0.9940 - val_loss: 0.0451 - val_accuracy: 0.9960 - 79ms/epoch - 5ms/step\n", "Epoch 292/400\n", "16/16 - 0s - loss: 0.0391 - accuracy: 0.9940 - val_loss: 0.0449 - val_accuracy: 0.9960 - 68ms/epoch - 4ms/step\n", "Epoch 293/400\n", "16/16 - 0s - loss: 0.0389 - accuracy: 0.9940 - val_loss: 0.0446 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 294/400\n", "16/16 - 0s - loss: 0.0386 - accuracy: 0.9940 - val_loss: 0.0444 - val_accuracy: 0.9960 - 111ms/epoch - 7ms/step\n", "Epoch 295/400\n", "16/16 - 0s - loss: 0.0385 - accuracy: 0.9940 - val_loss: 0.0442 - val_accuracy: 0.9960 - 116ms/epoch - 7ms/step\n", "Epoch 296/400\n", "16/16 - 0s - loss: 0.0382 - accuracy: 0.9940 - val_loss: 0.0439 - val_accuracy: 0.9960 - 76ms/epoch - 5ms/step\n", "Epoch 297/400\n", "16/16 - 0s - loss: 0.0380 - accuracy: 0.9940 - val_loss: 0.0436 - val_accuracy: 0.9960 - 69ms/epoch - 4ms/step\n", "Epoch 298/400\n", "16/16 - 0s - loss: 0.0378 - accuracy: 0.9940 - val_loss: 0.0434 - val_accuracy: 0.9960 - 69ms/epoch - 4ms/step\n", "Epoch 299/400\n", "16/16 - 0s - loss: 0.0376 - accuracy: 0.9940 - val_loss: 0.0432 - val_accuracy: 0.9960 - 115ms/epoch - 7ms/step\n", "Epoch 300/400\n", "16/16 - 0s - loss: 0.0374 - accuracy: 0.9940 - val_loss: 0.0429 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 301/400\n", "16/16 - 0s - loss: 0.0372 - accuracy: 0.9940 - val_loss: 0.0427 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 302/400\n", "16/16 - 0s - loss: 0.0370 - accuracy: 0.9940 - val_loss: 0.0424 - val_accuracy: 0.9980 - 68ms/epoch - 4ms/step\n", "Epoch 303/400\n", "16/16 - 0s - loss: 0.0368 - accuracy: 0.9940 - val_loss: 0.0422 - val_accuracy: 0.9980 - 79ms/epoch - 5ms/step\n", "Epoch 304/400\n", "16/16 - 0s - loss: 0.0366 - accuracy: 0.9940 - val_loss: 0.0420 - val_accuracy: 0.9980 - 113ms/epoch - 7ms/step\n", "Epoch 305/400\n", "16/16 - 0s - loss: 0.0364 - accuracy: 0.9940 - val_loss: 0.0417 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 306/400\n", "16/16 - 0s - loss: 0.0362 - accuracy: 0.9940 - val_loss: 0.0415 - val_accuracy: 0.9980 - 111ms/epoch - 7ms/step\n", "Epoch 307/400\n", "16/16 - 0s - loss: 0.0360 - accuracy: 0.9940 - val_loss: 0.0413 - val_accuracy: 0.9980 - 78ms/epoch - 5ms/step\n", "Epoch 308/400\n", "16/16 - 0s - loss: 0.0358 - accuracy: 0.9940 - val_loss: 0.0411 - val_accuracy: 0.9980 - 71ms/epoch - 4ms/step\n", "Epoch 309/400\n", "16/16 - 0s - loss: 0.0356 - accuracy: 0.9940 - val_loss: 0.0409 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 310/400\n", "16/16 - 0s - loss: 0.0354 - accuracy: 0.9940 - val_loss: 0.0406 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 311/400\n", "16/16 - 0s - loss: 0.0352 - accuracy: 0.9940 - val_loss: 0.0403 - val_accuracy: 0.9980 - 73ms/epoch - 5ms/step\n", "Epoch 312/400\n", "16/16 - 0s - loss: 0.0350 - accuracy: 0.9940 - val_loss: 0.0401 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 313/400\n", "16/16 - 0s - loss: 0.0348 - accuracy: 0.9940 - val_loss: 0.0399 - val_accuracy: 0.9980 - 120ms/epoch - 8ms/step\n", "Epoch 314/400\n", "16/16 - 0s - loss: 0.0346 - accuracy: 0.9940 - val_loss: 0.0397 - val_accuracy: 0.9980 - 80ms/epoch - 5ms/step\n", "Epoch 315/400\n", "16/16 - 0s - loss: 0.0344 - accuracy: 0.9940 - val_loss: 0.0394 - val_accuracy: 0.9980 - 72ms/epoch - 4ms/step\n", "Epoch 316/400\n", "16/16 - 0s - loss: 0.0342 - accuracy: 0.9940 - val_loss: 0.0392 - val_accuracy: 0.9980 - 71ms/epoch - 4ms/step\n", "Epoch 317/400\n", "16/16 - 0s - loss: 0.0340 - accuracy: 0.9940 - val_loss: 0.0390 - val_accuracy: 0.9980 - 117ms/epoch - 7ms/step\n", "Epoch 318/400\n", "16/16 - 0s - loss: 0.0338 - accuracy: 0.9940 - val_loss: 0.0387 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 319/400\n", "16/16 - 0s - loss: 0.0336 - accuracy: 0.9940 - val_loss: 0.0386 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 320/400\n", "16/16 - 0s - loss: 0.0334 - accuracy: 0.9940 - val_loss: 0.0383 - val_accuracy: 0.9980 - 69ms/epoch - 4ms/step\n", "Epoch 321/400\n", "16/16 - 0s - loss: 0.0332 - accuracy: 0.9940 - val_loss: 0.0381 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 322/400\n", "16/16 - 0s - loss: 0.0330 - accuracy: 0.9940 - val_loss: 0.0379 - val_accuracy: 0.9980 - 124ms/epoch - 8ms/step\n", "Epoch 323/400\n", "16/16 - 0s - loss: 0.0328 - accuracy: 0.9940 - val_loss: 0.0376 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 324/400\n", "16/16 - 0s - loss: 0.0326 - accuracy: 0.9940 - val_loss: 0.0374 - val_accuracy: 0.9980 - 71ms/epoch - 4ms/step\n", "Epoch 325/400\n", "16/16 - 0s - loss: 0.0324 - accuracy: 0.9940 - val_loss: 0.0372 - val_accuracy: 0.9980 - 113ms/epoch - 7ms/step\n", "Epoch 326/400\n", "16/16 - 0s - loss: 0.0322 - accuracy: 0.9940 - val_loss: 0.0370 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 327/400\n", "16/16 - 0s - loss: 0.0320 - accuracy: 0.9940 - val_loss: 0.0368 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 328/400\n", "16/16 - 0s - loss: 0.0319 - accuracy: 0.9940 - val_loss: 0.0366 - val_accuracy: 0.9980 - 117ms/epoch - 7ms/step\n", "Epoch 329/400\n", "16/16 - 0s - loss: 0.0317 - accuracy: 0.9940 - val_loss: 0.0364 - val_accuracy: 0.9980 - 115ms/epoch - 7ms/step\n", "Epoch 330/400\n", "16/16 - 0s - loss: 0.0315 - accuracy: 0.9940 - val_loss: 0.0361 - val_accuracy: 0.9980 - 114ms/epoch - 7ms/step\n", "Epoch 331/400\n", "16/16 - 0s - loss: 0.0313 - accuracy: 0.9940 - val_loss: 0.0359 - val_accuracy: 0.9980 - 111ms/epoch - 7ms/step\n", "Epoch 332/400\n", "16/16 - 0s - loss: 0.0311 - accuracy: 0.9940 - val_loss: 0.0357 - val_accuracy: 0.9980 - 120ms/epoch - 7ms/step\n", "Epoch 333/400\n", "16/16 - 0s - loss: 0.0309 - accuracy: 0.9960 - val_loss: 0.0355 - val_accuracy: 0.9980 - 112ms/epoch - 7ms/step\n", "Epoch 334/400\n", "16/16 - 0s - loss: 0.0307 - accuracy: 0.9960 - val_loss: 0.0353 - val_accuracy: 0.9980 - 76ms/epoch - 5ms/step\n", "Epoch 335/400\n", "16/16 - 0s - loss: 0.0305 - accuracy: 0.9960 - val_loss: 0.0351 - val_accuracy: 0.9980 - 109ms/epoch - 7ms/step\n", "Epoch 336/400\n", "16/16 - 0s - loss: 0.0304 - accuracy: 0.9960 - val_loss: 0.0349 - val_accuracy: 0.9980 - 72ms/epoch - 4ms/step\n", "Epoch 337/400\n", "16/16 - 0s - loss: 0.0302 - accuracy: 0.9960 - val_loss: 0.0347 - val_accuracy: 0.9980 - 73ms/epoch - 5ms/step\n", "Epoch 338/400\n", "16/16 - 0s - loss: 0.0300 - accuracy: 0.9960 - val_loss: 0.0345 - val_accuracy: 0.9980 - 115ms/epoch - 7ms/step\n", "Epoch 339/400\n", "16/16 - 0s - loss: 0.0298 - accuracy: 0.9960 - val_loss: 0.0343 - val_accuracy: 0.9980 - 117ms/epoch - 7ms/step\n", "Epoch 340/400\n", "16/16 - 0s - loss: 0.0296 - accuracy: 0.9960 - val_loss: 0.0340 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 341/400\n", "16/16 - 0s - loss: 0.0295 - accuracy: 0.9960 - val_loss: 0.0338 - val_accuracy: 0.9980 - 71ms/epoch - 4ms/step\n", "Epoch 342/400\n", "16/16 - 0s - loss: 0.0293 - accuracy: 0.9960 - val_loss: 0.0336 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 343/400\n", "16/16 - 0s - loss: 0.0291 - accuracy: 0.9960 - val_loss: 0.0334 - val_accuracy: 0.9980 - 73ms/epoch - 5ms/step\n", "Epoch 344/400\n", "16/16 - 0s - loss: 0.0289 - accuracy: 0.9960 - val_loss: 0.0332 - val_accuracy: 0.9980 - 118ms/epoch - 7ms/step\n", "Epoch 345/400\n", "16/16 - 0s - loss: 0.0288 - accuracy: 0.9960 - val_loss: 0.0330 - val_accuracy: 0.9980 - 72ms/epoch - 4ms/step\n", "Epoch 346/400\n", "16/16 - 0s - loss: 0.0286 - accuracy: 0.9960 - val_loss: 0.0328 - val_accuracy: 0.9980 - 79ms/epoch - 5ms/step\n", "Epoch 347/400\n", "16/16 - 0s - loss: 0.0284 - accuracy: 0.9960 - val_loss: 0.0326 - val_accuracy: 0.9980 - 117ms/epoch - 7ms/step\n", "Epoch 348/400\n", "16/16 - 0s - loss: 0.0283 - accuracy: 0.9960 - val_loss: 0.0324 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 349/400\n", "16/16 - 0s - loss: 0.0281 - accuracy: 0.9960 - val_loss: 0.0322 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 350/400\n", "16/16 - 0s - loss: 0.0279 - accuracy: 0.9960 - val_loss: 0.0321 - val_accuracy: 0.9980 - 117ms/epoch - 7ms/step\n", "Epoch 351/400\n", "16/16 - 0s - loss: 0.0278 - accuracy: 0.9960 - val_loss: 0.0319 - val_accuracy: 0.9980 - 68ms/epoch - 4ms/step\n", "Epoch 352/400\n", "16/16 - 0s - loss: 0.0276 - accuracy: 0.9960 - val_loss: 0.0317 - val_accuracy: 0.9980 - 112ms/epoch - 7ms/step\n", "Epoch 353/400\n", "16/16 - 0s - loss: 0.0274 - accuracy: 0.9960 - val_loss: 0.0315 - val_accuracy: 0.9980 - 112ms/epoch - 7ms/step\n", "Epoch 354/400\n", "16/16 - 0s - loss: 0.0273 - accuracy: 0.9960 - val_loss: 0.0314 - val_accuracy: 0.9980 - 68ms/epoch - 4ms/step\n", "Epoch 355/400\n", "16/16 - 0s - loss: 0.0271 - accuracy: 0.9960 - val_loss: 0.0312 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 356/400\n", "16/16 - 0s - loss: 0.0269 - accuracy: 0.9960 - val_loss: 0.0310 - val_accuracy: 0.9980 - 109ms/epoch - 7ms/step\n", "Epoch 357/400\n", "16/16 - 0s - loss: 0.0268 - accuracy: 0.9960 - val_loss: 0.0308 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 358/400\n", "16/16 - 0s - loss: 0.0266 - accuracy: 0.9960 - val_loss: 0.0307 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 359/400\n", "16/16 - 0s - loss: 0.0265 - accuracy: 0.9960 - val_loss: 0.0305 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 360/400\n", "16/16 - 0s - loss: 0.0263 - accuracy: 0.9960 - val_loss: 0.0303 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 361/400\n", "16/16 - 0s - loss: 0.0262 - accuracy: 0.9960 - val_loss: 0.0302 - val_accuracy: 1.0000 - 79ms/epoch - 5ms/step\n", "Epoch 362/400\n", "16/16 - 0s - loss: 0.0260 - accuracy: 0.9960 - val_loss: 0.0300 - val_accuracy: 1.0000 - 116ms/epoch - 7ms/step\n", "Epoch 363/400\n", "16/16 - 0s - loss: 0.0259 - accuracy: 0.9960 - val_loss: 0.0298 - val_accuracy: 1.0000 - 73ms/epoch - 5ms/step\n", "Epoch 364/400\n", "16/16 - 0s - loss: 0.0257 - accuracy: 0.9960 - val_loss: 0.0296 - val_accuracy: 1.0000 - 109ms/epoch - 7ms/step\n", "Epoch 365/400\n", "16/16 - 0s - loss: 0.0256 - accuracy: 0.9980 - val_loss: 0.0295 - val_accuracy: 1.0000 - 113ms/epoch - 7ms/step\n", "Epoch 366/400\n", "16/16 - 0s - loss: 0.0254 - accuracy: 0.9980 - val_loss: 0.0293 - val_accuracy: 1.0000 - 78ms/epoch - 5ms/step\n", "Epoch 367/400\n", "16/16 - 0s - loss: 0.0253 - accuracy: 0.9980 - val_loss: 0.0291 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 368/400\n", "16/16 - 0s - loss: 0.0251 - accuracy: 0.9980 - val_loss: 0.0290 - val_accuracy: 1.0000 - 122ms/epoch - 8ms/step\n", "Epoch 369/400\n", "16/16 - 0s - loss: 0.0250 - accuracy: 0.9980 - val_loss: 0.0288 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 370/400\n", "16/16 - 0s - loss: 0.0248 - accuracy: 0.9980 - val_loss: 0.0286 - val_accuracy: 1.0000 - 110ms/epoch - 7ms/step\n", "Epoch 371/400\n", "16/16 - 0s - loss: 0.0247 - accuracy: 0.9980 - val_loss: 0.0285 - val_accuracy: 1.0000 - 110ms/epoch - 7ms/step\n", "Epoch 372/400\n", "16/16 - 0s - loss: 0.0245 - accuracy: 0.9980 - val_loss: 0.0283 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 373/400\n", "16/16 - 0s - loss: 0.0244 - accuracy: 0.9980 - val_loss: 0.0282 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 374/400\n", "16/16 - 0s - loss: 0.0243 - accuracy: 0.9980 - val_loss: 0.0280 - val_accuracy: 1.0000 - 67ms/epoch - 4ms/step\n", "Epoch 375/400\n", "16/16 - 0s - loss: 0.0241 - accuracy: 0.9980 - val_loss: 0.0279 - val_accuracy: 1.0000 - 72ms/epoch - 4ms/step\n", "Epoch 376/400\n", "16/16 - 0s - loss: 0.0240 - accuracy: 0.9980 - val_loss: 0.0277 - val_accuracy: 1.0000 - 111ms/epoch - 7ms/step\n", "Epoch 377/400\n", "16/16 - 0s - loss: 0.0238 - accuracy: 0.9980 - val_loss: 0.0276 - val_accuracy: 1.0000 - 79ms/epoch - 5ms/step\n", "Epoch 378/400\n", "16/16 - 0s - loss: 0.0237 - accuracy: 0.9980 - val_loss: 0.0275 - val_accuracy: 1.0000 - 72ms/epoch - 4ms/step\n", "Epoch 379/400\n", "16/16 - 0s - loss: 0.0236 - accuracy: 0.9980 - val_loss: 0.0273 - val_accuracy: 1.0000 - 89ms/epoch - 6ms/step\n", "Epoch 380/400\n", "16/16 - 0s - loss: 0.0234 - accuracy: 0.9980 - val_loss: 0.0272 - val_accuracy: 1.0000 - 114ms/epoch - 7ms/step\n", "Epoch 381/400\n", "16/16 - 0s - loss: 0.0233 - accuracy: 0.9980 - val_loss: 0.0271 - val_accuracy: 1.0000 - 81ms/epoch - 5ms/step\n", "Epoch 382/400\n", "16/16 - 0s - loss: 0.0232 - accuracy: 0.9980 - val_loss: 0.0270 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 383/400\n", "16/16 - 0s - loss: 0.0230 - accuracy: 0.9980 - val_loss: 0.0268 - val_accuracy: 1.0000 - 109ms/epoch - 7ms/step\n", "Epoch 384/400\n", "16/16 - 0s - loss: 0.0229 - accuracy: 0.9980 - val_loss: 0.0267 - val_accuracy: 1.0000 - 117ms/epoch - 7ms/step\n", "Epoch 385/400\n", "16/16 - 0s - loss: 0.0227 - accuracy: 0.9980 - val_loss: 0.0266 - val_accuracy: 1.0000 - 109ms/epoch - 7ms/step\n", "Epoch 386/400\n", "16/16 - 0s - loss: 0.0226 - accuracy: 0.9980 - val_loss: 0.0264 - val_accuracy: 1.0000 - 68ms/epoch - 4ms/step\n", "Epoch 387/400\n", "16/16 - 0s - loss: 0.0225 - accuracy: 0.9980 - val_loss: 0.0263 - val_accuracy: 1.0000 - 73ms/epoch - 5ms/step\n", "Epoch 388/400\n", "16/16 - 0s - loss: 0.0223 - accuracy: 0.9980 - val_loss: 0.0262 - val_accuracy: 1.0000 - 111ms/epoch - 7ms/step\n", "Epoch 389/400\n", "16/16 - 0s - loss: 0.0222 - accuracy: 0.9980 - val_loss: 0.0261 - val_accuracy: 1.0000 - 91ms/epoch - 6ms/step\n", "Epoch 390/400\n", "16/16 - 0s - loss: 0.0221 - accuracy: 0.9980 - val_loss: 0.0260 - val_accuracy: 1.0000 - 116ms/epoch - 7ms/step\n", "Epoch 391/400\n", "16/16 - 0s - loss: 0.0220 - accuracy: 0.9980 - val_loss: 0.0259 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 392/400\n", "16/16 - 0s - loss: 0.0218 - accuracy: 0.9980 - val_loss: 0.0257 - val_accuracy: 1.0000 - 73ms/epoch - 5ms/step\n", "Epoch 393/400\n", "16/16 - 0s - loss: 0.0217 - accuracy: 0.9980 - val_loss: 0.0256 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 394/400\n", "16/16 - 0s - loss: 0.0216 - accuracy: 0.9980 - val_loss: 0.0255 - val_accuracy: 1.0000 - 114ms/epoch - 7ms/step\n", "Epoch 395/400\n", "16/16 - 0s - loss: 0.0215 - accuracy: 0.9980 - val_loss: 0.0253 - val_accuracy: 1.0000 - 74ms/epoch - 5ms/step\n", "Epoch 396/400\n", "16/16 - 0s - loss: 0.0213 - accuracy: 0.9980 - val_loss: 0.0252 - val_accuracy: 1.0000 - 119ms/epoch - 7ms/step\n", "Epoch 397/400\n", "16/16 - 0s - loss: 0.0212 - accuracy: 0.9980 - val_loss: 0.0251 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 398/400\n", "16/16 - 0s - loss: 0.0211 - accuracy: 0.9980 - val_loss: 0.0250 - val_accuracy: 1.0000 - 114ms/epoch - 7ms/step\n", "Epoch 399/400\n", "16/16 - 0s - loss: 0.0210 - accuracy: 0.9980 - val_loss: 0.0248 - val_accuracy: 1.0000 - 112ms/epoch - 7ms/step\n", "Epoch 400/400\n", "16/16 - 0s - loss: 0.0209 - accuracy: 0.9980 - val_loss: 0.0247 - val_accuracy: 1.0000 - 69ms/epoch - 4ms/step\n", "第3个弱分类器训练完毕\n", "Epoch 1/400\n", "16/16 - 1s - loss: 0.3360 - accuracy: 0.8420 - val_loss: 0.3349 - val_accuracy: 0.8280 - 880ms/epoch - 55ms/step\n", "Epoch 2/400\n", "16/16 - 0s - loss: 0.3268 - accuracy: 0.8480 - val_loss: 0.3262 - val_accuracy: 0.8340 - 69ms/epoch - 4ms/step\n", "Epoch 3/400\n", "16/16 - 0s - loss: 0.3183 - accuracy: 0.8520 - val_loss: 0.3177 - val_accuracy: 0.8400 - 74ms/epoch - 5ms/step\n", "Epoch 4/400\n", "16/16 - 0s - loss: 0.3096 - accuracy: 0.8580 - val_loss: 0.3102 - val_accuracy: 0.8440 - 110ms/epoch - 7ms/step\n", "Epoch 5/400\n", "16/16 - 0s - loss: 0.3022 - accuracy: 0.8620 - val_loss: 0.3027 - val_accuracy: 0.8540 - 68ms/epoch - 4ms/step\n", "Epoch 6/400\n", "16/16 - 0s - loss: 0.2944 - accuracy: 0.8680 - val_loss: 0.2959 - val_accuracy: 0.8560 - 109ms/epoch - 7ms/step\n", "Epoch 7/400\n", "16/16 - 0s - loss: 0.2875 - accuracy: 0.8720 - val_loss: 0.2895 - val_accuracy: 0.8580 - 75ms/epoch - 5ms/step\n", "Epoch 8/400\n", "16/16 - 0s - loss: 0.2809 - accuracy: 0.8780 - val_loss: 0.2832 - val_accuracy: 0.8660 - 72ms/epoch - 5ms/step\n", "Epoch 9/400\n", "16/16 - 0s - loss: 0.2744 - accuracy: 0.8900 - val_loss: 0.2776 - val_accuracy: 0.8760 - 117ms/epoch - 7ms/step\n", "Epoch 10/400\n", "16/16 - 0s - loss: 0.2687 - accuracy: 0.8960 - val_loss: 0.2719 - val_accuracy: 0.8820 - 72ms/epoch - 5ms/step\n", "Epoch 11/400\n", "16/16 - 0s - loss: 0.2631 - accuracy: 0.9000 - val_loss: 0.2668 - val_accuracy: 0.8920 - 70ms/epoch - 4ms/step\n", "Epoch 12/400\n", "16/16 - 0s - loss: 0.2577 - accuracy: 0.9040 - val_loss: 0.2621 - val_accuracy: 0.9000 - 122ms/epoch - 8ms/step\n", "Epoch 13/400\n", "16/16 - 0s - loss: 0.2528 - accuracy: 0.9100 - val_loss: 0.2577 - val_accuracy: 0.9060 - 109ms/epoch - 7ms/step\n", "Epoch 14/400\n", "16/16 - 0s - loss: 0.2481 - accuracy: 0.9140 - val_loss: 0.2536 - val_accuracy: 0.9100 - 114ms/epoch - 7ms/step\n", "Epoch 15/400\n", "16/16 - 0s - loss: 0.2440 - accuracy: 0.9140 - val_loss: 0.2495 - val_accuracy: 0.9160 - 110ms/epoch - 7ms/step\n", "Epoch 16/400\n", "16/16 - 0s - loss: 0.2397 - accuracy: 0.9180 - val_loss: 0.2459 - val_accuracy: 0.9180 - 69ms/epoch - 4ms/step\n", "Epoch 17/400\n", "16/16 - 0s - loss: 0.2361 - accuracy: 0.9200 - val_loss: 0.2423 - val_accuracy: 0.9260 - 68ms/epoch - 4ms/step\n", "Epoch 18/400\n", "16/16 - 0s - loss: 0.2323 - accuracy: 0.9320 - val_loss: 0.2391 - val_accuracy: 0.9260 - 73ms/epoch - 5ms/step\n", "Epoch 19/400\n", "16/16 - 0s - loss: 0.2289 - accuracy: 0.9360 - val_loss: 0.2362 - val_accuracy: 0.9260 - 115ms/epoch - 7ms/step\n", "Epoch 20/400\n", "16/16 - 0s - loss: 0.2257 - accuracy: 0.9380 - val_loss: 0.2334 - val_accuracy: 0.9300 - 75ms/epoch - 5ms/step\n", "Epoch 21/400\n", "16/16 - 0s - loss: 0.2227 - accuracy: 0.9440 - val_loss: 0.2307 - val_accuracy: 0.9320 - 74ms/epoch - 5ms/step\n", "Epoch 22/400\n", "16/16 - 0s - loss: 0.2198 - accuracy: 0.9440 - val_loss: 0.2281 - val_accuracy: 0.9380 - 75ms/epoch - 5ms/step\n", "Epoch 23/400\n", "16/16 - 0s - loss: 0.2170 - accuracy: 0.9500 - val_loss: 0.2258 - val_accuracy: 0.9380 - 87ms/epoch - 5ms/step\n", "Epoch 24/400\n", "16/16 - 0s - loss: 0.2145 - accuracy: 0.9500 - val_loss: 0.2236 - val_accuracy: 0.9380 - 114ms/epoch - 7ms/step\n", "Epoch 25/400\n", "16/16 - 0s - loss: 0.2121 - accuracy: 0.9520 - val_loss: 0.2215 - val_accuracy: 0.9400 - 74ms/epoch - 5ms/step\n", "Epoch 26/400\n", "16/16 - 0s - loss: 0.2097 - accuracy: 0.9520 - val_loss: 0.2197 - val_accuracy: 0.9420 - 121ms/epoch - 8ms/step\n", "Epoch 27/400\n", "16/16 - 0s - loss: 0.2077 - accuracy: 0.9520 - val_loss: 0.2176 - val_accuracy: 0.9420 - 75ms/epoch - 5ms/step\n", "Epoch 28/400\n", "16/16 - 0s - loss: 0.2055 - accuracy: 0.9540 - val_loss: 0.2158 - val_accuracy: 0.9440 - 115ms/epoch - 7ms/step\n", "Epoch 29/400\n", "16/16 - 0s - loss: 0.2035 - accuracy: 0.9540 - val_loss: 0.2141 - val_accuracy: 0.9480 - 112ms/epoch - 7ms/step\n", "Epoch 30/400\n", "16/16 - 0s - loss: 0.2016 - accuracy: 0.9560 - val_loss: 0.2125 - val_accuracy: 0.9480 - 117ms/epoch - 7ms/step\n", "Epoch 31/400\n", "16/16 - 0s - loss: 0.1997 - accuracy: 0.9600 - val_loss: 0.2109 - val_accuracy: 0.9480 - 118ms/epoch - 7ms/step\n", "Epoch 32/400\n", "16/16 - 0s - loss: 0.1979 - accuracy: 0.9600 - val_loss: 0.2094 - val_accuracy: 0.9500 - 77ms/epoch - 5ms/step\n", "Epoch 33/400\n", "16/16 - 0s - loss: 0.1962 - accuracy: 0.9640 - val_loss: 0.2080 - val_accuracy: 0.9500 - 80ms/epoch - 5ms/step\n", "Epoch 34/400\n", "16/16 - 0s - loss: 0.1945 - accuracy: 0.9640 - val_loss: 0.2066 - val_accuracy: 0.9560 - 74ms/epoch - 5ms/step\n", "Epoch 35/400\n", "16/16 - 0s - loss: 0.1930 - accuracy: 0.9660 - val_loss: 0.2051 - val_accuracy: 0.9560 - 116ms/epoch - 7ms/step\n", "Epoch 36/400\n", "16/16 - 0s - loss: 0.1913 - accuracy: 0.9680 - val_loss: 0.2037 - val_accuracy: 0.9620 - 110ms/epoch - 7ms/step\n", "Epoch 37/400\n", "16/16 - 0s - loss: 0.1897 - accuracy: 0.9700 - val_loss: 0.2024 - val_accuracy: 0.9660 - 119ms/epoch - 7ms/step\n", "Epoch 38/400\n", "16/16 - 0s - loss: 0.1881 - accuracy: 0.9720 - val_loss: 0.2011 - val_accuracy: 0.9660 - 115ms/epoch - 7ms/step\n", "Epoch 39/400\n", "16/16 - 0s - loss: 0.1866 - accuracy: 0.9720 - val_loss: 0.1998 - val_accuracy: 0.9680 - 116ms/epoch - 7ms/step\n", "Epoch 40/400\n", "16/16 - 0s - loss: 0.1851 - accuracy: 0.9720 - val_loss: 0.1986 - val_accuracy: 0.9680 - 116ms/epoch - 7ms/step\n", "Epoch 41/400\n", "16/16 - 0s - loss: 0.1837 - accuracy: 0.9720 - val_loss: 0.1973 - val_accuracy: 0.9700 - 68ms/epoch - 4ms/step\n", "Epoch 42/400\n", "16/16 - 0s - loss: 0.1822 - accuracy: 0.9720 - val_loss: 0.1961 - val_accuracy: 0.9700 - 118ms/epoch - 7ms/step\n", "Epoch 43/400\n", "16/16 - 0s - loss: 0.1809 - accuracy: 0.9720 - val_loss: 0.1949 - val_accuracy: 0.9700 - 114ms/epoch - 7ms/step\n", "Epoch 44/400\n", "16/16 - 0s - loss: 0.1794 - accuracy: 0.9740 - val_loss: 0.1936 - val_accuracy: 0.9700 - 71ms/epoch - 4ms/step\n", "Epoch 45/400\n", "16/16 - 0s - loss: 0.1781 - accuracy: 0.9760 - val_loss: 0.1924 - val_accuracy: 0.9720 - 80ms/epoch - 5ms/step\n", "Epoch 46/400\n", "16/16 - 0s - loss: 0.1767 - accuracy: 0.9760 - val_loss: 0.1912 - val_accuracy: 0.9740 - 71ms/epoch - 4ms/step\n", "Epoch 47/400\n", "16/16 - 0s - loss: 0.1753 - accuracy: 0.9760 - val_loss: 0.1900 - val_accuracy: 0.9740 - 74ms/epoch - 5ms/step\n", "Epoch 48/400\n", "16/16 - 0s - loss: 0.1740 - accuracy: 0.9740 - val_loss: 0.1888 - val_accuracy: 0.9740 - 116ms/epoch - 7ms/step\n", "Epoch 49/400\n", "16/16 - 0s - loss: 0.1726 - accuracy: 0.9760 - val_loss: 0.1876 - val_accuracy: 0.9740 - 117ms/epoch - 7ms/step\n", "Epoch 50/400\n", "16/16 - 0s - loss: 0.1713 - accuracy: 0.9760 - val_loss: 0.1865 - val_accuracy: 0.9740 - 117ms/epoch - 7ms/step\n", "Epoch 51/400\n", "16/16 - 0s - loss: 0.1700 - accuracy: 0.9780 - val_loss: 0.1853 - val_accuracy: 0.9740 - 76ms/epoch - 5ms/step\n", "Epoch 52/400\n", "16/16 - 0s - loss: 0.1687 - accuracy: 0.9780 - val_loss: 0.1841 - val_accuracy: 0.9740 - 80ms/epoch - 5ms/step\n", "Epoch 53/400\n", "16/16 - 0s - loss: 0.1673 - accuracy: 0.9780 - val_loss: 0.1830 - val_accuracy: 0.9740 - 81ms/epoch - 5ms/step\n", "Epoch 54/400\n", "16/16 - 0s - loss: 0.1660 - accuracy: 0.9800 - val_loss: 0.1818 - val_accuracy: 0.9740 - 84ms/epoch - 5ms/step\n", "Epoch 55/400\n", "16/16 - 0s - loss: 0.1647 - accuracy: 0.9800 - val_loss: 0.1807 - val_accuracy: 0.9760 - 117ms/epoch - 7ms/step\n", "Epoch 56/400\n", "16/16 - 0s - loss: 0.1635 - accuracy: 0.9820 - val_loss: 0.1795 - val_accuracy: 0.9760 - 109ms/epoch - 7ms/step\n", "Epoch 57/400\n", "16/16 - 0s - loss: 0.1622 - accuracy: 0.9800 - val_loss: 0.1784 - val_accuracy: 0.9760 - 112ms/epoch - 7ms/step\n", "Epoch 58/400\n", "16/16 - 0s - loss: 0.1609 - accuracy: 0.9840 - val_loss: 0.1772 - val_accuracy: 0.9760 - 77ms/epoch - 5ms/step\n", "Epoch 59/400\n", "16/16 - 0s - loss: 0.1596 - accuracy: 0.9840 - val_loss: 0.1760 - val_accuracy: 0.9760 - 110ms/epoch - 7ms/step\n", "Epoch 60/400\n", "16/16 - 0s - loss: 0.1584 - accuracy: 0.9840 - val_loss: 0.1749 - val_accuracy: 0.9760 - 81ms/epoch - 5ms/step\n", "Epoch 61/400\n", "16/16 - 0s - loss: 0.1571 - accuracy: 0.9840 - val_loss: 0.1737 - val_accuracy: 0.9760 - 75ms/epoch - 5ms/step\n", "Epoch 62/400\n", "16/16 - 0s - loss: 0.1558 - accuracy: 0.9840 - val_loss: 0.1726 - val_accuracy: 0.9760 - 70ms/epoch - 4ms/step\n", "Epoch 63/400\n", "16/16 - 0s - loss: 0.1546 - accuracy: 0.9840 - val_loss: 0.1714 - val_accuracy: 0.9760 - 76ms/epoch - 5ms/step\n", "Epoch 64/400\n", "16/16 - 0s - loss: 0.1533 - accuracy: 0.9840 - val_loss: 0.1703 - val_accuracy: 0.9760 - 74ms/epoch - 5ms/step\n", "Epoch 65/400\n", "16/16 - 0s - loss: 0.1521 - accuracy: 0.9840 - val_loss: 0.1691 - val_accuracy: 0.9760 - 74ms/epoch - 5ms/step\n", "Epoch 66/400\n", "16/16 - 0s - loss: 0.1509 - accuracy: 0.9840 - val_loss: 0.1679 - val_accuracy: 0.9760 - 112ms/epoch - 7ms/step\n", "Epoch 67/400\n", "16/16 - 0s - loss: 0.1497 - accuracy: 0.9840 - val_loss: 0.1668 - val_accuracy: 0.9760 - 75ms/epoch - 5ms/step\n", "Epoch 68/400\n", "16/16 - 0s - loss: 0.1485 - accuracy: 0.9840 - val_loss: 0.1656 - val_accuracy: 0.9760 - 81ms/epoch - 5ms/step\n", "Epoch 69/400\n", "16/16 - 0s - loss: 0.1472 - accuracy: 0.9840 - val_loss: 0.1644 - val_accuracy: 0.9760 - 110ms/epoch - 7ms/step\n", "Epoch 70/400\n", "16/16 - 0s - loss: 0.1460 - accuracy: 0.9840 - val_loss: 0.1633 - val_accuracy: 0.9760 - 109ms/epoch - 7ms/step\n", "Epoch 71/400\n", "16/16 - 0s - loss: 0.1448 - accuracy: 0.9840 - val_loss: 0.1621 - val_accuracy: 0.9760 - 113ms/epoch - 7ms/step\n", "Epoch 72/400\n", "16/16 - 0s - loss: 0.1436 - accuracy: 0.9840 - val_loss: 0.1610 - val_accuracy: 0.9760 - 75ms/epoch - 5ms/step\n", "Epoch 73/400\n", "16/16 - 0s - loss: 0.1424 - accuracy: 0.9840 - val_loss: 0.1598 - val_accuracy: 0.9760 - 73ms/epoch - 5ms/step\n", "Epoch 74/400\n", "16/16 - 0s - loss: 0.1413 - accuracy: 0.9840 - val_loss: 0.1587 - val_accuracy: 0.9760 - 120ms/epoch - 7ms/step\n", "Epoch 75/400\n", "16/16 - 0s - loss: 0.1401 - accuracy: 0.9840 - val_loss: 0.1576 - val_accuracy: 0.9760 - 111ms/epoch - 7ms/step\n", "Epoch 76/400\n", "16/16 - 0s - loss: 0.1389 - accuracy: 0.9840 - val_loss: 0.1564 - val_accuracy: 0.9760 - 76ms/epoch - 5ms/step\n", "Epoch 77/400\n", "16/16 - 0s - loss: 0.1377 - accuracy: 0.9840 - val_loss: 0.1553 - val_accuracy: 0.9760 - 74ms/epoch - 5ms/step\n", "Epoch 78/400\n", "16/16 - 0s - loss: 0.1366 - accuracy: 0.9840 - val_loss: 0.1542 - val_accuracy: 0.9760 - 72ms/epoch - 4ms/step\n", "Epoch 79/400\n", "16/16 - 0s - loss: 0.1354 - accuracy: 0.9840 - val_loss: 0.1530 - val_accuracy: 0.9760 - 79ms/epoch - 5ms/step\n", "Epoch 80/400\n", "16/16 - 0s - loss: 0.1343 - accuracy: 0.9840 - val_loss: 0.1519 - val_accuracy: 0.9760 - 113ms/epoch - 7ms/step\n", "Epoch 81/400\n", "16/16 - 0s - loss: 0.1332 - accuracy: 0.9840 - val_loss: 0.1508 - val_accuracy: 0.9760 - 71ms/epoch - 4ms/step\n", "Epoch 82/400\n", "16/16 - 0s - loss: 0.1320 - accuracy: 0.9840 - val_loss: 0.1497 - val_accuracy: 0.9760 - 116ms/epoch - 7ms/step\n", "Epoch 83/400\n", "16/16 - 0s - loss: 0.1309 - accuracy: 0.9840 - val_loss: 0.1486 - val_accuracy: 0.9760 - 76ms/epoch - 5ms/step\n", "Epoch 84/400\n", "16/16 - 0s - loss: 0.1298 - accuracy: 0.9840 - val_loss: 0.1475 - val_accuracy: 0.9780 - 115ms/epoch - 7ms/step\n", "Epoch 85/400\n", "16/16 - 0s - loss: 0.1287 - accuracy: 0.9840 - val_loss: 0.1463 - val_accuracy: 0.9780 - 76ms/epoch - 5ms/step\n", "Epoch 86/400\n", "16/16 - 0s - loss: 0.1276 - accuracy: 0.9840 - val_loss: 0.1453 - val_accuracy: 0.9780 - 80ms/epoch - 5ms/step\n", "Epoch 87/400\n", "16/16 - 0s - loss: 0.1265 - accuracy: 0.9840 - val_loss: 0.1442 - val_accuracy: 0.9780 - 78ms/epoch - 5ms/step\n", "Epoch 88/400\n", "16/16 - 0s - loss: 0.1254 - accuracy: 0.9840 - val_loss: 0.1431 - val_accuracy: 0.9780 - 90ms/epoch - 6ms/step\n", "Epoch 89/400\n", "16/16 - 0s - loss: 0.1243 - accuracy: 0.9840 - val_loss: 0.1420 - val_accuracy: 0.9780 - 77ms/epoch - 5ms/step\n", "Epoch 90/400\n", "16/16 - 0s - loss: 0.1233 - accuracy: 0.9840 - val_loss: 0.1409 - val_accuracy: 0.9780 - 90ms/epoch - 6ms/step\n", "Epoch 91/400\n", "16/16 - 0s - loss: 0.1222 - accuracy: 0.9840 - val_loss: 0.1398 - val_accuracy: 0.9780 - 112ms/epoch - 7ms/step\n", "Epoch 92/400\n", "16/16 - 0s - loss: 0.1211 - accuracy: 0.9840 - val_loss: 0.1388 - val_accuracy: 0.9780 - 111ms/epoch - 7ms/step\n", "Epoch 93/400\n", "16/16 - 0s - loss: 0.1201 - accuracy: 0.9840 - val_loss: 0.1377 - val_accuracy: 0.9780 - 74ms/epoch - 5ms/step\n", "Epoch 94/400\n", "16/16 - 0s - loss: 0.1190 - accuracy: 0.9860 - val_loss: 0.1366 - val_accuracy: 0.9780 - 75ms/epoch - 5ms/step\n", "Epoch 95/400\n", "16/16 - 0s - loss: 0.1180 - accuracy: 0.9860 - val_loss: 0.1356 - val_accuracy: 0.9780 - 113ms/epoch - 7ms/step\n", "Epoch 96/400\n", "16/16 - 0s - loss: 0.1170 - accuracy: 0.9860 - val_loss: 0.1345 - val_accuracy: 0.9780 - 74ms/epoch - 5ms/step\n", "Epoch 97/400\n", "16/16 - 0s - loss: 0.1160 - accuracy: 0.9880 - val_loss: 0.1335 - val_accuracy: 0.9780 - 112ms/epoch - 7ms/step\n", "Epoch 98/400\n", "16/16 - 0s - loss: 0.1149 - accuracy: 0.9880 - val_loss: 0.1324 - val_accuracy: 0.9800 - 114ms/epoch - 7ms/step\n", "Epoch 99/400\n", "16/16 - 0s - loss: 0.1140 - accuracy: 0.9880 - val_loss: 0.1313 - val_accuracy: 0.9800 - 113ms/epoch - 7ms/step\n", "Epoch 100/400\n", "16/16 - 0s - loss: 0.1129 - accuracy: 0.9880 - val_loss: 0.1303 - val_accuracy: 0.9800 - 74ms/epoch - 5ms/step\n", "Epoch 101/400\n", "16/16 - 0s - loss: 0.1120 - accuracy: 0.9880 - val_loss: 0.1293 - val_accuracy: 0.9800 - 111ms/epoch - 7ms/step\n", "Epoch 102/400\n", "16/16 - 0s - loss: 0.1110 - accuracy: 0.9900 - val_loss: 0.1283 - val_accuracy: 0.9800 - 70ms/epoch - 4ms/step\n", "Epoch 103/400\n", "16/16 - 0s - loss: 0.1100 - accuracy: 0.9900 - val_loss: 0.1272 - val_accuracy: 0.9800 - 110ms/epoch - 7ms/step\n", "Epoch 104/400\n", "16/16 - 0s - loss: 0.1090 - accuracy: 0.9900 - val_loss: 0.1263 - val_accuracy: 0.9800 - 116ms/epoch - 7ms/step\n", "Epoch 105/400\n", "16/16 - 0s - loss: 0.1081 - accuracy: 0.9900 - val_loss: 0.1252 - val_accuracy: 0.9800 - 80ms/epoch - 5ms/step\n", "Epoch 106/400\n", "16/16 - 0s - loss: 0.1071 - accuracy: 0.9920 - val_loss: 0.1243 - val_accuracy: 0.9820 - 110ms/epoch - 7ms/step\n", "Epoch 107/400\n", "16/16 - 0s - loss: 0.1062 - accuracy: 0.9920 - val_loss: 0.1233 - val_accuracy: 0.9840 - 78ms/epoch - 5ms/step\n", "Epoch 108/400\n", "16/16 - 0s - loss: 0.1053 - accuracy: 0.9940 - val_loss: 0.1224 - val_accuracy: 0.9840 - 80ms/epoch - 5ms/step\n", "Epoch 109/400\n", "16/16 - 0s - loss: 0.1044 - accuracy: 0.9940 - val_loss: 0.1214 - val_accuracy: 0.9840 - 68ms/epoch - 4ms/step\n", "Epoch 110/400\n", "16/16 - 0s - loss: 0.1035 - accuracy: 0.9940 - val_loss: 0.1204 - val_accuracy: 0.9840 - 112ms/epoch - 7ms/step\n", "Epoch 111/400\n", "16/16 - 0s - loss: 0.1026 - accuracy: 0.9940 - val_loss: 0.1195 - val_accuracy: 0.9840 - 77ms/epoch - 5ms/step\n", "Epoch 112/400\n", "16/16 - 0s - loss: 0.1017 - accuracy: 0.9940 - val_loss: 0.1186 - val_accuracy: 0.9840 - 113ms/epoch - 7ms/step\n", "Epoch 113/400\n", "16/16 - 0s - loss: 0.1008 - accuracy: 0.9940 - val_loss: 0.1176 - val_accuracy: 0.9840 - 73ms/epoch - 5ms/step\n", "Epoch 114/400\n", "16/16 - 0s - loss: 0.1000 - accuracy: 0.9940 - val_loss: 0.1168 - val_accuracy: 0.9840 - 74ms/epoch - 5ms/step\n", "Epoch 115/400\n", "16/16 - 0s - loss: 0.0991 - accuracy: 0.9940 - val_loss: 0.1158 - val_accuracy: 0.9860 - 74ms/epoch - 5ms/step\n", "Epoch 116/400\n", "16/16 - 0s - loss: 0.0983 - accuracy: 0.9940 - val_loss: 0.1149 - val_accuracy: 0.9860 - 79ms/epoch - 5ms/step\n", "Epoch 117/400\n", "16/16 - 0s - loss: 0.0975 - accuracy: 0.9940 - val_loss: 0.1141 - val_accuracy: 0.9860 - 80ms/epoch - 5ms/step\n", "Epoch 118/400\n", "16/16 - 0s - loss: 0.0967 - accuracy: 0.9940 - val_loss: 0.1132 - val_accuracy: 0.9860 - 74ms/epoch - 5ms/step\n", "Epoch 119/400\n", "16/16 - 0s - loss: 0.0958 - accuracy: 0.9940 - val_loss: 0.1124 - val_accuracy: 0.9860 - 119ms/epoch - 7ms/step\n", "Epoch 120/400\n", "16/16 - 0s - loss: 0.0951 - accuracy: 0.9940 - val_loss: 0.1114 - val_accuracy: 0.9860 - 113ms/epoch - 7ms/step\n", "Epoch 121/400\n", "16/16 - 0s - loss: 0.0942 - accuracy: 0.9940 - val_loss: 0.1107 - val_accuracy: 0.9860 - 116ms/epoch - 7ms/step\n", "Epoch 122/400\n", "16/16 - 0s - loss: 0.0935 - accuracy: 0.9940 - val_loss: 0.1098 - val_accuracy: 0.9860 - 74ms/epoch - 5ms/step\n", "Epoch 123/400\n", "16/16 - 0s - loss: 0.0927 - accuracy: 0.9940 - val_loss: 0.1090 - val_accuracy: 0.9860 - 109ms/epoch - 7ms/step\n", "Epoch 124/400\n", "16/16 - 0s - loss: 0.0919 - accuracy: 0.9940 - val_loss: 0.1081 - val_accuracy: 0.9860 - 115ms/epoch - 7ms/step\n", "Epoch 125/400\n", "16/16 - 0s - loss: 0.0912 - accuracy: 0.9940 - val_loss: 0.1073 - val_accuracy: 0.9860 - 71ms/epoch - 4ms/step\n", "Epoch 126/400\n", "16/16 - 0s - loss: 0.0904 - accuracy: 0.9940 - val_loss: 0.1065 - val_accuracy: 0.9860 - 82ms/epoch - 5ms/step\n", "Epoch 127/400\n", "16/16 - 0s - loss: 0.0897 - accuracy: 0.9940 - val_loss: 0.1057 - val_accuracy: 0.9860 - 110ms/epoch - 7ms/step\n", "Epoch 128/400\n", "16/16 - 0s - loss: 0.0890 - accuracy: 0.9940 - val_loss: 0.1049 - val_accuracy: 0.9860 - 72ms/epoch - 5ms/step\n", "Epoch 129/400\n", "16/16 - 0s - loss: 0.0882 - accuracy: 0.9940 - val_loss: 0.1042 - val_accuracy: 0.9860 - 112ms/epoch - 7ms/step\n", "Epoch 130/400\n", "16/16 - 0s - loss: 0.0875 - accuracy: 0.9940 - val_loss: 0.1034 - val_accuracy: 0.9860 - 82ms/epoch - 5ms/step\n", "Epoch 131/400\n", "16/16 - 0s - loss: 0.0868 - accuracy: 0.9940 - val_loss: 0.1027 - val_accuracy: 0.9880 - 73ms/epoch - 5ms/step\n", "Epoch 132/400\n", "16/16 - 0s - loss: 0.0861 - accuracy: 0.9940 - val_loss: 0.1019 - val_accuracy: 0.9880 - 111ms/epoch - 7ms/step\n", "Epoch 133/400\n", "16/16 - 0s - loss: 0.0854 - accuracy: 0.9940 - val_loss: 0.1012 - val_accuracy: 0.9880 - 116ms/epoch - 7ms/step\n", "Epoch 134/400\n", "16/16 - 0s - loss: 0.0847 - accuracy: 0.9940 - val_loss: 0.1005 - val_accuracy: 0.9880 - 72ms/epoch - 4ms/step\n", "Epoch 135/400\n", "16/16 - 0s - loss: 0.0841 - accuracy: 0.9940 - val_loss: 0.0998 - val_accuracy: 0.9880 - 86ms/epoch - 5ms/step\n", "Epoch 136/400\n", "16/16 - 0s - loss: 0.0834 - accuracy: 0.9940 - val_loss: 0.0991 - val_accuracy: 0.9880 - 74ms/epoch - 5ms/step\n", "Epoch 137/400\n", "16/16 - 0s - loss: 0.0828 - accuracy: 0.9940 - val_loss: 0.0984 - val_accuracy: 0.9880 - 121ms/epoch - 8ms/step\n", "Epoch 138/400\n", "16/16 - 0s - loss: 0.0821 - accuracy: 0.9940 - val_loss: 0.0977 - val_accuracy: 0.9880 - 72ms/epoch - 5ms/step\n", "Epoch 139/400\n", "16/16 - 0s - loss: 0.0815 - accuracy: 0.9940 - val_loss: 0.0970 - val_accuracy: 0.9880 - 72ms/epoch - 4ms/step\n", "Epoch 140/400\n", "16/16 - 0s - loss: 0.0809 - accuracy: 0.9940 - val_loss: 0.0963 - val_accuracy: 0.9880 - 75ms/epoch - 5ms/step\n", "Epoch 141/400\n", "16/16 - 0s - loss: 0.0803 - accuracy: 0.9940 - val_loss: 0.0956 - val_accuracy: 0.9880 - 72ms/epoch - 4ms/step\n", "Epoch 142/400\n", "16/16 - 0s - loss: 0.0797 - accuracy: 0.9940 - val_loss: 0.0950 - val_accuracy: 0.9880 - 79ms/epoch - 5ms/step\n", "Epoch 143/400\n", "16/16 - 0s - loss: 0.0791 - accuracy: 0.9940 - val_loss: 0.0943 - val_accuracy: 0.9880 - 112ms/epoch - 7ms/step\n", "Epoch 144/400\n", "16/16 - 0s - loss: 0.0785 - accuracy: 0.9940 - val_loss: 0.0936 - val_accuracy: 0.9900 - 111ms/epoch - 7ms/step\n", "Epoch 145/400\n", "16/16 - 0s - loss: 0.0779 - accuracy: 0.9940 - val_loss: 0.0930 - val_accuracy: 0.9900 - 71ms/epoch - 4ms/step\n", "Epoch 146/400\n", "16/16 - 0s - loss: 0.0773 - accuracy: 0.9940 - val_loss: 0.0924 - val_accuracy: 0.9900 - 72ms/epoch - 4ms/step\n", "Epoch 147/400\n", "16/16 - 0s - loss: 0.0768 - accuracy: 0.9940 - val_loss: 0.0917 - val_accuracy: 0.9920 - 117ms/epoch - 7ms/step\n", "Epoch 148/400\n", "16/16 - 0s - loss: 0.0762 - accuracy: 0.9940 - val_loss: 0.0911 - val_accuracy: 0.9920 - 97ms/epoch - 6ms/step\n", "Epoch 149/400\n", "16/16 - 0s - loss: 0.0756 - accuracy: 0.9940 - val_loss: 0.0905 - val_accuracy: 0.9920 - 77ms/epoch - 5ms/step\n", "Epoch 150/400\n", "16/16 - 0s - loss: 0.0751 - accuracy: 0.9940 - val_loss: 0.0899 - val_accuracy: 0.9920 - 73ms/epoch - 5ms/step\n", "Epoch 151/400\n", "16/16 - 0s - loss: 0.0745 - accuracy: 0.9940 - val_loss: 0.0893 - val_accuracy: 0.9920 - 118ms/epoch - 7ms/step\n", "Epoch 152/400\n", "16/16 - 0s - loss: 0.0740 - accuracy: 0.9940 - val_loss: 0.0887 - val_accuracy: 0.9920 - 72ms/epoch - 5ms/step\n", "Epoch 153/400\n", "16/16 - 0s - loss: 0.0735 - accuracy: 0.9940 - val_loss: 0.0881 - val_accuracy: 0.9920 - 75ms/epoch - 5ms/step\n", "Epoch 154/400\n", "16/16 - 0s - loss: 0.0730 - accuracy: 0.9940 - val_loss: 0.0876 - val_accuracy: 0.9920 - 75ms/epoch - 5ms/step\n", "Epoch 155/400\n", "16/16 - 0s - loss: 0.0724 - accuracy: 0.9940 - val_loss: 0.0869 - val_accuracy: 0.9920 - 72ms/epoch - 4ms/step\n", "Epoch 156/400\n", "16/16 - 0s - loss: 0.0719 - accuracy: 0.9960 - val_loss: 0.0863 - val_accuracy: 0.9920 - 72ms/epoch - 4ms/step\n", "Epoch 157/400\n", "16/16 - 0s - loss: 0.0714 - accuracy: 0.9960 - val_loss: 0.0858 - val_accuracy: 0.9920 - 72ms/epoch - 4ms/step\n", "Epoch 158/400\n", "16/16 - 0s - loss: 0.0709 - accuracy: 0.9960 - val_loss: 0.0852 - val_accuracy: 0.9920 - 115ms/epoch - 7ms/step\n", "Epoch 159/400\n", "16/16 - 0s - loss: 0.0704 - accuracy: 0.9960 - val_loss: 0.0846 - val_accuracy: 0.9920 - 91ms/epoch - 6ms/step\n", "Epoch 160/400\n", "16/16 - 0s - loss: 0.0700 - accuracy: 0.9960 - val_loss: 0.0841 - val_accuracy: 0.9920 - 119ms/epoch - 7ms/step\n", "Epoch 161/400\n", "16/16 - 0s - loss: 0.0695 - accuracy: 0.9960 - val_loss: 0.0835 - val_accuracy: 0.9940 - 78ms/epoch - 5ms/step\n", "Epoch 162/400\n", "16/16 - 0s - loss: 0.0690 - accuracy: 0.9960 - val_loss: 0.0830 - val_accuracy: 0.9940 - 117ms/epoch - 7ms/step\n", "Epoch 163/400\n", "16/16 - 0s - loss: 0.0685 - accuracy: 0.9960 - val_loss: 0.0825 - val_accuracy: 0.9940 - 74ms/epoch - 5ms/step\n", "Epoch 164/400\n", "16/16 - 0s - loss: 0.0681 - accuracy: 0.9960 - val_loss: 0.0820 - val_accuracy: 0.9940 - 78ms/epoch - 5ms/step\n", "Epoch 165/400\n", "16/16 - 0s - loss: 0.0676 - accuracy: 0.9960 - val_loss: 0.0814 - val_accuracy: 0.9940 - 75ms/epoch - 5ms/step\n", "Epoch 166/400\n", "16/16 - 0s - loss: 0.0671 - accuracy: 0.9960 - val_loss: 0.0809 - val_accuracy: 0.9940 - 74ms/epoch - 5ms/step\n", "Epoch 167/400\n", "16/16 - 0s - loss: 0.0667 - accuracy: 0.9960 - val_loss: 0.0805 - val_accuracy: 0.9940 - 75ms/epoch - 5ms/step\n", "Epoch 168/400\n", "16/16 - 0s - loss: 0.0663 - accuracy: 0.9960 - val_loss: 0.0800 - val_accuracy: 0.9960 - 75ms/epoch - 5ms/step\n", "Epoch 169/400\n", "16/16 - 0s - loss: 0.0658 - accuracy: 0.9960 - val_loss: 0.0795 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 170/400\n", "16/16 - 0s - loss: 0.0654 - accuracy: 0.9960 - val_loss: 0.0790 - val_accuracy: 0.9960 - 67ms/epoch - 4ms/step\n", "Epoch 171/400\n", "16/16 - 0s - loss: 0.0650 - accuracy: 0.9960 - val_loss: 0.0785 - val_accuracy: 0.9960 - 125ms/epoch - 8ms/step\n", "Epoch 172/400\n", "16/16 - 0s - loss: 0.0645 - accuracy: 0.9960 - val_loss: 0.0780 - val_accuracy: 0.9960 - 112ms/epoch - 7ms/step\n", "Epoch 173/400\n", "16/16 - 0s - loss: 0.0641 - accuracy: 0.9960 - val_loss: 0.0775 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 174/400\n", "16/16 - 0s - loss: 0.0637 - accuracy: 0.9960 - val_loss: 0.0770 - val_accuracy: 0.9960 - 118ms/epoch - 7ms/step\n", "Epoch 175/400\n", "16/16 - 0s - loss: 0.0633 - accuracy: 0.9960 - val_loss: 0.0766 - val_accuracy: 0.9960 - 75ms/epoch - 5ms/step\n", "Epoch 176/400\n", "16/16 - 0s - loss: 0.0629 - accuracy: 0.9960 - val_loss: 0.0761 - val_accuracy: 0.9960 - 112ms/epoch - 7ms/step\n", "Epoch 177/400\n", "16/16 - 0s - loss: 0.0625 - accuracy: 0.9960 - val_loss: 0.0756 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 178/400\n", "16/16 - 0s - loss: 0.0621 - accuracy: 0.9960 - val_loss: 0.0752 - val_accuracy: 0.9960 - 71ms/epoch - 4ms/step\n", "Epoch 179/400\n", "16/16 - 0s - loss: 0.0617 - accuracy: 0.9960 - val_loss: 0.0747 - val_accuracy: 0.9960 - 113ms/epoch - 7ms/step\n", "Epoch 180/400\n", "16/16 - 0s - loss: 0.0614 - accuracy: 0.9960 - val_loss: 0.0743 - val_accuracy: 0.9960 - 114ms/epoch - 7ms/step\n", "Epoch 181/400\n", "16/16 - 0s - loss: 0.0610 - accuracy: 0.9960 - val_loss: 0.0738 - val_accuracy: 0.9960 - 84ms/epoch - 5ms/step\n", "Epoch 182/400\n", "16/16 - 0s - loss: 0.0606 - accuracy: 0.9960 - val_loss: 0.0733 - val_accuracy: 0.9960 - 114ms/epoch - 7ms/step\n", "Epoch 183/400\n", "16/16 - 0s - loss: 0.0602 - accuracy: 0.9960 - val_loss: 0.0729 - val_accuracy: 0.9960 - 114ms/epoch - 7ms/step\n", "Epoch 184/400\n", "16/16 - 0s - loss: 0.0598 - accuracy: 0.9960 - val_loss: 0.0724 - val_accuracy: 0.9960 - 115ms/epoch - 7ms/step\n", "Epoch 185/400\n", "16/16 - 0s - loss: 0.0595 - accuracy: 0.9960 - val_loss: 0.0721 - val_accuracy: 0.9960 - 75ms/epoch - 5ms/step\n", "Epoch 186/400\n", "16/16 - 0s - loss: 0.0591 - accuracy: 0.9960 - val_loss: 0.0717 - val_accuracy: 0.9960 - 72ms/epoch - 4ms/step\n", "Epoch 187/400\n", "16/16 - 0s - loss: 0.0587 - accuracy: 0.9960 - val_loss: 0.0712 - val_accuracy: 0.9960 - 119ms/epoch - 7ms/step\n", "Epoch 188/400\n", "16/16 - 0s - loss: 0.0584 - accuracy: 0.9960 - val_loss: 0.0708 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 189/400\n", "16/16 - 0s - loss: 0.0580 - accuracy: 0.9960 - val_loss: 0.0704 - val_accuracy: 0.9960 - 76ms/epoch - 5ms/step\n", "Epoch 190/400\n", "16/16 - 0s - loss: 0.0577 - accuracy: 0.9960 - val_loss: 0.0700 - val_accuracy: 0.9960 - 75ms/epoch - 5ms/step\n", "Epoch 191/400\n", "16/16 - 0s - loss: 0.0573 - accuracy: 0.9960 - val_loss: 0.0696 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 192/400\n", "16/16 - 0s - loss: 0.0570 - accuracy: 0.9960 - val_loss: 0.0692 - val_accuracy: 0.9960 - 77ms/epoch - 5ms/step\n", "Epoch 193/400\n", "16/16 - 0s - loss: 0.0566 - accuracy: 0.9960 - val_loss: 0.0688 - val_accuracy: 0.9960 - 114ms/epoch - 7ms/step\n", "Epoch 194/400\n", "16/16 - 0s - loss: 0.0563 - accuracy: 0.9960 - val_loss: 0.0684 - val_accuracy: 0.9960 - 75ms/epoch - 5ms/step\n", "Epoch 195/400\n", "16/16 - 0s - loss: 0.0560 - accuracy: 0.9960 - val_loss: 0.0681 - val_accuracy: 0.9960 - 78ms/epoch - 5ms/step\n", "Epoch 196/400\n", "16/16 - 0s - loss: 0.0556 - accuracy: 0.9960 - val_loss: 0.0676 - val_accuracy: 0.9960 - 74ms/epoch - 5ms/step\n", "Epoch 197/400\n", "16/16 - 0s - loss: 0.0553 - accuracy: 0.9960 - val_loss: 0.0672 - val_accuracy: 0.9960 - 112ms/epoch - 7ms/step\n", "Epoch 198/400\n", "16/16 - 0s - loss: 0.0550 - accuracy: 0.9960 - val_loss: 0.0668 - val_accuracy: 0.9960 - 112ms/epoch - 7ms/step\n", "Epoch 199/400\n", "16/16 - 0s - loss: 0.0547 - accuracy: 0.9960 - val_loss: 0.0664 - val_accuracy: 0.9960 - 117ms/epoch - 7ms/step\n", "Epoch 200/400\n", "16/16 - 0s - loss: 0.0544 - accuracy: 0.9960 - val_loss: 0.0661 - val_accuracy: 0.9960 - 115ms/epoch - 7ms/step\n", "Epoch 201/400\n", "16/16 - 0s - loss: 0.0540 - accuracy: 0.9960 - val_loss: 0.0658 - val_accuracy: 0.9960 - 115ms/epoch - 7ms/step\n", "Epoch 202/400\n", "16/16 - 0s - loss: 0.0537 - accuracy: 0.9960 - val_loss: 0.0654 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 203/400\n", "16/16 - 0s - loss: 0.0534 - accuracy: 0.9960 - val_loss: 0.0650 - val_accuracy: 0.9960 - 70ms/epoch - 4ms/step\n", "Epoch 204/400\n", "16/16 - 0s - loss: 0.0531 - accuracy: 0.9960 - val_loss: 0.0645 - val_accuracy: 0.9960 - 118ms/epoch - 7ms/step\n", "Epoch 205/400\n", "16/16 - 0s - loss: 0.0528 - accuracy: 0.9960 - val_loss: 0.0643 - val_accuracy: 0.9960 - 114ms/epoch - 7ms/step\n", "Epoch 206/400\n", "16/16 - 0s - loss: 0.0525 - accuracy: 0.9960 - val_loss: 0.0639 - val_accuracy: 0.9960 - 70ms/epoch - 4ms/step\n", "Epoch 207/400\n", "16/16 - 0s - loss: 0.0522 - accuracy: 0.9980 - val_loss: 0.0635 - val_accuracy: 0.9960 - 70ms/epoch - 4ms/step\n", "Epoch 208/400\n", "16/16 - 0s - loss: 0.0519 - accuracy: 0.9980 - val_loss: 0.0631 - val_accuracy: 0.9980 - 118ms/epoch - 7ms/step\n", "Epoch 209/400\n", "16/16 - 0s - loss: 0.0516 - accuracy: 1.0000 - val_loss: 0.0628 - val_accuracy: 0.9980 - 119ms/epoch - 7ms/step\n", "Epoch 210/400\n", "16/16 - 0s - loss: 0.0513 - accuracy: 1.0000 - val_loss: 0.0624 - val_accuracy: 0.9980 - 83ms/epoch - 5ms/step\n", "Epoch 211/400\n", "16/16 - 0s - loss: 0.0510 - accuracy: 1.0000 - val_loss: 0.0621 - val_accuracy: 0.9980 - 83ms/epoch - 5ms/step\n", "Epoch 212/400\n", "16/16 - 0s - loss: 0.0508 - accuracy: 1.0000 - val_loss: 0.0618 - val_accuracy: 0.9980 - 100ms/epoch - 6ms/step\n", "Epoch 213/400\n", "16/16 - 0s - loss: 0.0505 - accuracy: 1.0000 - val_loss: 0.0615 - val_accuracy: 0.9980 - 111ms/epoch - 7ms/step\n", "Epoch 214/400\n", "16/16 - 0s - loss: 0.0502 - accuracy: 1.0000 - val_loss: 0.0611 - val_accuracy: 0.9980 - 127ms/epoch - 8ms/step\n", "Epoch 215/400\n", "16/16 - 0s - loss: 0.0499 - accuracy: 1.0000 - val_loss: 0.0607 - val_accuracy: 0.9980 - 112ms/epoch - 7ms/step\n", "Epoch 216/400\n", "16/16 - 0s - loss: 0.0496 - accuracy: 1.0000 - val_loss: 0.0605 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 217/400\n", "16/16 - 0s - loss: 0.0494 - accuracy: 1.0000 - val_loss: 0.0601 - val_accuracy: 0.9980 - 73ms/epoch - 5ms/step\n", "Epoch 218/400\n", "16/16 - 0s - loss: 0.0491 - accuracy: 1.0000 - val_loss: 0.0598 - val_accuracy: 0.9980 - 111ms/epoch - 7ms/step\n", "Epoch 219/400\n", "16/16 - 0s - loss: 0.0488 - accuracy: 1.0000 - val_loss: 0.0595 - val_accuracy: 0.9980 - 85ms/epoch - 5ms/step\n", "Epoch 220/400\n", "16/16 - 0s - loss: 0.0486 - accuracy: 1.0000 - val_loss: 0.0592 - val_accuracy: 0.9980 - 115ms/epoch - 7ms/step\n", "Epoch 221/400\n", "16/16 - 0s - loss: 0.0483 - accuracy: 1.0000 - val_loss: 0.0588 - val_accuracy: 0.9980 - 117ms/epoch - 7ms/step\n", "Epoch 222/400\n", "16/16 - 0s - loss: 0.0480 - accuracy: 1.0000 - val_loss: 0.0585 - val_accuracy: 0.9980 - 115ms/epoch - 7ms/step\n", "Epoch 223/400\n", "16/16 - 0s - loss: 0.0478 - accuracy: 1.0000 - val_loss: 0.0582 - val_accuracy: 0.9980 - 111ms/epoch - 7ms/step\n", "Epoch 224/400\n", "16/16 - 0s - loss: 0.0475 - accuracy: 1.0000 - val_loss: 0.0579 - val_accuracy: 0.9980 - 77ms/epoch - 5ms/step\n", "Epoch 225/400\n", "16/16 - 0s - loss: 0.0473 - accuracy: 1.0000 - val_loss: 0.0575 - val_accuracy: 0.9980 - 78ms/epoch - 5ms/step\n", "Epoch 226/400\n", "16/16 - 0s - loss: 0.0470 - accuracy: 1.0000 - val_loss: 0.0572 - val_accuracy: 0.9980 - 79ms/epoch - 5ms/step\n", "Epoch 227/400\n", "16/16 - 0s - loss: 0.0468 - accuracy: 1.0000 - val_loss: 0.0570 - val_accuracy: 0.9980 - 111ms/epoch - 7ms/step\n", "Epoch 228/400\n", "16/16 - 0s - loss: 0.0465 - accuracy: 1.0000 - val_loss: 0.0567 - val_accuracy: 0.9980 - 116ms/epoch - 7ms/step\n", "Epoch 229/400\n", "16/16 - 0s - loss: 0.0463 - accuracy: 1.0000 - val_loss: 0.0564 - val_accuracy: 0.9980 - 115ms/epoch - 7ms/step\n", "Epoch 230/400\n", "16/16 - 0s - loss: 0.0460 - accuracy: 1.0000 - val_loss: 0.0561 - val_accuracy: 0.9980 - 79ms/epoch - 5ms/step\n", "Epoch 231/400\n", "16/16 - 0s - loss: 0.0458 - accuracy: 1.0000 - val_loss: 0.0558 - val_accuracy: 0.9980 - 121ms/epoch - 8ms/step\n", "Epoch 232/400\n", "16/16 - 0s - loss: 0.0455 - accuracy: 1.0000 - val_loss: 0.0554 - val_accuracy: 0.9980 - 119ms/epoch - 7ms/step\n", "Epoch 233/400\n", "16/16 - 0s - loss: 0.0453 - accuracy: 1.0000 - val_loss: 0.0552 - val_accuracy: 0.9980 - 118ms/epoch - 7ms/step\n", "Epoch 234/400\n", "16/16 - 0s - loss: 0.0450 - accuracy: 1.0000 - val_loss: 0.0548 - val_accuracy: 0.9980 - 115ms/epoch - 7ms/step\n", "Epoch 235/400\n", "16/16 - 0s - loss: 0.0448 - accuracy: 1.0000 - val_loss: 0.0546 - val_accuracy: 0.9980 - 125ms/epoch - 8ms/step\n", "Epoch 236/400\n", "16/16 - 0s - loss: 0.0446 - accuracy: 1.0000 - val_loss: 0.0543 - val_accuracy: 0.9980 - 123ms/epoch - 8ms/step\n", "Epoch 237/400\n", "16/16 - 0s - loss: 0.0443 - accuracy: 1.0000 - val_loss: 0.0540 - val_accuracy: 0.9980 - 111ms/epoch - 7ms/step\n", "Epoch 238/400\n", "16/16 - 0s - loss: 0.0441 - accuracy: 1.0000 - val_loss: 0.0537 - val_accuracy: 0.9980 - 116ms/epoch - 7ms/step\n", "Epoch 239/400\n", "16/16 - 0s - loss: 0.0439 - accuracy: 1.0000 - val_loss: 0.0535 - val_accuracy: 0.9980 - 80ms/epoch - 5ms/step\n", "Epoch 240/400\n", "16/16 - 0s - loss: 0.0437 - accuracy: 1.0000 - val_loss: 0.0533 - val_accuracy: 0.9980 - 84ms/epoch - 5ms/step\n", "Epoch 241/400\n", "16/16 - 0s - loss: 0.0434 - accuracy: 1.0000 - val_loss: 0.0530 - val_accuracy: 0.9980 - 79ms/epoch - 5ms/step\n", "Epoch 242/400\n", "16/16 - 0s - loss: 0.0432 - accuracy: 1.0000 - val_loss: 0.0527 - val_accuracy: 0.9980 - 116ms/epoch - 7ms/step\n", "Epoch 243/400\n", "16/16 - 0s - loss: 0.0430 - accuracy: 1.0000 - val_loss: 0.0525 - val_accuracy: 0.9980 - 82ms/epoch - 5ms/step\n", "Epoch 244/400\n", "16/16 - 0s - loss: 0.0428 - accuracy: 1.0000 - val_loss: 0.0522 - val_accuracy: 0.9980 - 118ms/epoch - 7ms/step\n", "Epoch 245/400\n", "16/16 - 0s - loss: 0.0426 - accuracy: 1.0000 - val_loss: 0.0520 - val_accuracy: 0.9980 - 76ms/epoch - 5ms/step\n", "Epoch 246/400\n", "16/16 - 0s - loss: 0.0423 - accuracy: 1.0000 - val_loss: 0.0516 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 247/400\n", "16/16 - 0s - loss: 0.0421 - accuracy: 1.0000 - val_loss: 0.0515 - val_accuracy: 0.9980 - 76ms/epoch - 5ms/step\n", "Epoch 248/400\n", "16/16 - 0s - loss: 0.0419 - accuracy: 1.0000 - val_loss: 0.0512 - val_accuracy: 0.9980 - 73ms/epoch - 5ms/step\n", "Epoch 249/400\n", "16/16 - 0s - loss: 0.0417 - accuracy: 1.0000 - val_loss: 0.0509 - val_accuracy: 0.9980 - 73ms/epoch - 5ms/step\n", "Epoch 250/400\n", "16/16 - 0s - loss: 0.0415 - accuracy: 1.0000 - val_loss: 0.0506 - val_accuracy: 0.9980 - 72ms/epoch - 5ms/step\n", "Epoch 251/400\n", "16/16 - 0s - loss: 0.0413 - accuracy: 1.0000 - val_loss: 0.0503 - val_accuracy: 0.9980 - 111ms/epoch - 7ms/step\n", "Epoch 252/400\n", "16/16 - 0s - loss: 0.0410 - accuracy: 1.0000 - val_loss: 0.0501 - val_accuracy: 0.9980 - 72ms/epoch - 5ms/step\n", "Epoch 253/400\n", "16/16 - 0s - loss: 0.0408 - accuracy: 1.0000 - val_loss: 0.0499 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 254/400\n", "16/16 - 0s - loss: 0.0406 - accuracy: 1.0000 - val_loss: 0.0496 - val_accuracy: 0.9980 - 112ms/epoch - 7ms/step\n", "Epoch 255/400\n", "16/16 - 0s - loss: 0.0404 - accuracy: 1.0000 - val_loss: 0.0494 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 256/400\n", "16/16 - 0s - loss: 0.0402 - accuracy: 1.0000 - val_loss: 0.0491 - val_accuracy: 0.9980 - 111ms/epoch - 7ms/step\n", "Epoch 257/400\n", "16/16 - 0s - loss: 0.0400 - accuracy: 1.0000 - val_loss: 0.0489 - val_accuracy: 0.9980 - 109ms/epoch - 7ms/step\n", "Epoch 258/400\n", "16/16 - 0s - loss: 0.0398 - accuracy: 1.0000 - val_loss: 0.0486 - val_accuracy: 0.9980 - 83ms/epoch - 5ms/step\n", "Epoch 259/400\n", "16/16 - 0s - loss: 0.0396 - accuracy: 1.0000 - val_loss: 0.0484 - val_accuracy: 0.9980 - 79ms/epoch - 5ms/step\n", "Epoch 260/400\n", "16/16 - 0s - loss: 0.0394 - accuracy: 1.0000 - val_loss: 0.0482 - val_accuracy: 0.9980 - 114ms/epoch - 7ms/step\n", "Epoch 261/400\n", "16/16 - 0s - loss: 0.0392 - accuracy: 1.0000 - val_loss: 0.0479 - val_accuracy: 0.9980 - 78ms/epoch - 5ms/step\n", "Epoch 262/400\n", "16/16 - 0s - loss: 0.0390 - accuracy: 1.0000 - val_loss: 0.0476 - val_accuracy: 0.9980 - 84ms/epoch - 5ms/step\n", "Epoch 263/400\n", "16/16 - 0s - loss: 0.0389 - accuracy: 1.0000 - val_loss: 0.0474 - val_accuracy: 0.9980 - 80ms/epoch - 5ms/step\n", "Epoch 264/400\n", "16/16 - 0s - loss: 0.0386 - accuracy: 1.0000 - val_loss: 0.0472 - val_accuracy: 0.9980 - 113ms/epoch - 7ms/step\n", "Epoch 265/400\n", "16/16 - 0s - loss: 0.0385 - accuracy: 1.0000 - val_loss: 0.0470 - val_accuracy: 0.9980 - 109ms/epoch - 7ms/step\n", "Epoch 266/400\n", "16/16 - 0s - loss: 0.0383 - accuracy: 1.0000 - val_loss: 0.0468 - val_accuracy: 0.9980 - 72ms/epoch - 4ms/step\n", "Epoch 267/400\n", "16/16 - 0s - loss: 0.0381 - accuracy: 1.0000 - val_loss: 0.0465 - val_accuracy: 0.9980 - 110ms/epoch - 7ms/step\n", "Epoch 268/400\n", "16/16 - 0s - loss: 0.0379 - accuracy: 1.0000 - val_loss: 0.0463 - val_accuracy: 0.9980 - 113ms/epoch - 7ms/step\n", "Epoch 269/400\n", "16/16 - 0s - loss: 0.0377 - accuracy: 1.0000 - val_loss: 0.0460 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 270/400\n", "16/16 - 0s - loss: 0.0375 - accuracy: 1.0000 - val_loss: 0.0458 - val_accuracy: 0.9980 - 72ms/epoch - 5ms/step\n", "Epoch 271/400\n", "16/16 - 0s - loss: 0.0373 - accuracy: 1.0000 - val_loss: 0.0456 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 272/400\n", "16/16 - 0s - loss: 0.0372 - accuracy: 1.0000 - val_loss: 0.0454 - val_accuracy: 0.9980 - 116ms/epoch - 7ms/step\n", "Epoch 273/400\n", "16/16 - 0s - loss: 0.0370 - accuracy: 1.0000 - val_loss: 0.0451 - val_accuracy: 0.9980 - 73ms/epoch - 5ms/step\n", "Epoch 274/400\n", "16/16 - 0s - loss: 0.0368 - accuracy: 1.0000 - val_loss: 0.0450 - val_accuracy: 0.9980 - 78ms/epoch - 5ms/step\n", "Epoch 275/400\n", "16/16 - 0s - loss: 0.0366 - accuracy: 1.0000 - val_loss: 0.0447 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 276/400\n", "16/16 - 0s - loss: 0.0364 - accuracy: 1.0000 - val_loss: 0.0445 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 277/400\n", "16/16 - 0s - loss: 0.0363 - accuracy: 1.0000 - val_loss: 0.0443 - val_accuracy: 0.9980 - 113ms/epoch - 7ms/step\n", "Epoch 278/400\n", "16/16 - 0s - loss: 0.0361 - accuracy: 1.0000 - val_loss: 0.0441 - val_accuracy: 0.9980 - 109ms/epoch - 7ms/step\n", "Epoch 279/400\n", "16/16 - 0s - loss: 0.0359 - accuracy: 1.0000 - val_loss: 0.0439 - val_accuracy: 0.9980 - 81ms/epoch - 5ms/step\n", "Epoch 280/400\n", "16/16 - 0s - loss: 0.0358 - accuracy: 1.0000 - val_loss: 0.0436 - val_accuracy: 0.9980 - 73ms/epoch - 5ms/step\n", "Epoch 281/400\n", "16/16 - 0s - loss: 0.0356 - accuracy: 1.0000 - val_loss: 0.0433 - val_accuracy: 0.9980 - 68ms/epoch - 4ms/step\n", "Epoch 282/400\n", "16/16 - 0s - loss: 0.0354 - accuracy: 1.0000 - val_loss: 0.0432 - val_accuracy: 0.9980 - 81ms/epoch - 5ms/step\n", "Epoch 283/400\n", "16/16 - 0s - loss: 0.0352 - accuracy: 1.0000 - val_loss: 0.0430 - val_accuracy: 0.9980 - 112ms/epoch - 7ms/step\n", "Epoch 284/400\n", "16/16 - 0s - loss: 0.0351 - accuracy: 1.0000 - val_loss: 0.0429 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 285/400\n", "16/16 - 0s - loss: 0.0349 - accuracy: 1.0000 - val_loss: 0.0426 - val_accuracy: 0.9980 - 72ms/epoch - 4ms/step\n", "Epoch 286/400\n", "16/16 - 0s - loss: 0.0347 - accuracy: 1.0000 - val_loss: 0.0424 - val_accuracy: 0.9980 - 76ms/epoch - 5ms/step\n", "Epoch 287/400\n", "16/16 - 0s - loss: 0.0346 - accuracy: 1.0000 - val_loss: 0.0422 - val_accuracy: 0.9980 - 73ms/epoch - 5ms/step\n", "Epoch 288/400\n", "16/16 - 0s - loss: 0.0344 - accuracy: 1.0000 - val_loss: 0.0421 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 289/400\n", "16/16 - 0s - loss: 0.0343 - accuracy: 1.0000 - val_loss: 0.0418 - val_accuracy: 0.9980 - 115ms/epoch - 7ms/step\n", "Epoch 290/400\n", "16/16 - 0s - loss: 0.0341 - accuracy: 1.0000 - val_loss: 0.0416 - val_accuracy: 0.9980 - 113ms/epoch - 7ms/step\n", "Epoch 291/400\n", "16/16 - 0s - loss: 0.0339 - accuracy: 1.0000 - val_loss: 0.0415 - val_accuracy: 0.9980 - 113ms/epoch - 7ms/step\n", "Epoch 292/400\n", "16/16 - 0s - loss: 0.0338 - accuracy: 1.0000 - val_loss: 0.0412 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 293/400\n", "16/16 - 0s - loss: 0.0336 - accuracy: 1.0000 - val_loss: 0.0411 - val_accuracy: 0.9980 - 118ms/epoch - 7ms/step\n", "Epoch 294/400\n", "16/16 - 0s - loss: 0.0335 - accuracy: 1.0000 - val_loss: 0.0409 - val_accuracy: 0.9980 - 123ms/epoch - 8ms/step\n", "Epoch 295/400\n", "16/16 - 0s - loss: 0.0333 - accuracy: 1.0000 - val_loss: 0.0407 - val_accuracy: 0.9980 - 78ms/epoch - 5ms/step\n", "Epoch 296/400\n", "16/16 - 0s - loss: 0.0331 - accuracy: 1.0000 - val_loss: 0.0405 - val_accuracy: 0.9980 - 118ms/epoch - 7ms/step\n", "Epoch 297/400\n", "16/16 - 0s - loss: 0.0330 - accuracy: 1.0000 - val_loss: 0.0403 - val_accuracy: 0.9980 - 70ms/epoch - 4ms/step\n", "Epoch 298/400\n", "16/16 - 0s - loss: 0.0328 - accuracy: 1.0000 - val_loss: 0.0401 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 299/400\n", "16/16 - 0s - loss: 0.0327 - accuracy: 1.0000 - val_loss: 0.0399 - val_accuracy: 0.9980 - 123ms/epoch - 8ms/step\n", "Epoch 300/400\n", "16/16 - 0s - loss: 0.0325 - accuracy: 1.0000 - val_loss: 0.0397 - val_accuracy: 0.9980 - 72ms/epoch - 4ms/step\n", "Epoch 301/400\n", "16/16 - 0s - loss: 0.0324 - accuracy: 1.0000 - val_loss: 0.0395 - val_accuracy: 0.9980 - 114ms/epoch - 7ms/step\n", "Epoch 302/400\n", "16/16 - 0s - loss: 0.0322 - accuracy: 1.0000 - val_loss: 0.0393 - val_accuracy: 1.0000 - 124ms/epoch - 8ms/step\n", "Epoch 303/400\n", "16/16 - 0s - loss: 0.0321 - accuracy: 1.0000 - val_loss: 0.0392 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 304/400\n", "16/16 - 0s - loss: 0.0319 - accuracy: 1.0000 - val_loss: 0.0391 - val_accuracy: 1.0000 - 84ms/epoch - 5ms/step\n", "Epoch 305/400\n", "16/16 - 0s - loss: 0.0318 - accuracy: 1.0000 - val_loss: 0.0389 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 306/400\n", "16/16 - 0s - loss: 0.0316 - accuracy: 1.0000 - val_loss: 0.0387 - val_accuracy: 1.0000 - 72ms/epoch - 4ms/step\n", "Epoch 307/400\n", "16/16 - 0s - loss: 0.0315 - accuracy: 1.0000 - val_loss: 0.0384 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 308/400\n", "16/16 - 0s - loss: 0.0313 - accuracy: 1.0000 - val_loss: 0.0383 - val_accuracy: 1.0000 - 77ms/epoch - 5ms/step\n", "Epoch 309/400\n", "16/16 - 0s - loss: 0.0312 - accuracy: 1.0000 - val_loss: 0.0381 - val_accuracy: 1.0000 - 74ms/epoch - 5ms/step\n", "Epoch 310/400\n", "16/16 - 0s - loss: 0.0310 - accuracy: 1.0000 - val_loss: 0.0379 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 311/400\n", "16/16 - 0s - loss: 0.0309 - accuracy: 1.0000 - val_loss: 0.0378 - val_accuracy: 1.0000 - 112ms/epoch - 7ms/step\n", "Epoch 312/400\n", "16/16 - 0s - loss: 0.0308 - accuracy: 1.0000 - val_loss: 0.0376 - val_accuracy: 1.0000 - 117ms/epoch - 7ms/step\n", "Epoch 313/400\n", "16/16 - 0s - loss: 0.0306 - accuracy: 1.0000 - val_loss: 0.0375 - val_accuracy: 1.0000 - 120ms/epoch - 7ms/step\n", "Epoch 314/400\n", "16/16 - 0s - loss: 0.0305 - accuracy: 1.0000 - val_loss: 0.0373 - val_accuracy: 1.0000 - 112ms/epoch - 7ms/step\n", "Epoch 315/400\n", "16/16 - 0s - loss: 0.0304 - accuracy: 1.0000 - val_loss: 0.0371 - val_accuracy: 1.0000 - 89ms/epoch - 6ms/step\n", "Epoch 316/400\n", "16/16 - 0s - loss: 0.0302 - accuracy: 1.0000 - val_loss: 0.0369 - val_accuracy: 1.0000 - 79ms/epoch - 5ms/step\n", "Epoch 317/400\n", "16/16 - 0s - loss: 0.0301 - accuracy: 1.0000 - val_loss: 0.0368 - val_accuracy: 1.0000 - 113ms/epoch - 7ms/step\n", "Epoch 318/400\n", "16/16 - 0s - loss: 0.0299 - accuracy: 1.0000 - val_loss: 0.0366 - val_accuracy: 1.0000 - 79ms/epoch - 5ms/step\n", "Epoch 319/400\n", "16/16 - 0s - loss: 0.0298 - accuracy: 1.0000 - val_loss: 0.0365 - val_accuracy: 1.0000 - 115ms/epoch - 7ms/step\n", "Epoch 320/400\n", "16/16 - 0s - loss: 0.0297 - accuracy: 1.0000 - val_loss: 0.0363 - val_accuracy: 1.0000 - 77ms/epoch - 5ms/step\n", "Epoch 321/400\n", "16/16 - 0s - loss: 0.0295 - accuracy: 1.0000 - val_loss: 0.0361 - val_accuracy: 1.0000 - 74ms/epoch - 5ms/step\n", "Epoch 322/400\n", "16/16 - 0s - loss: 0.0294 - accuracy: 1.0000 - val_loss: 0.0359 - val_accuracy: 1.0000 - 109ms/epoch - 7ms/step\n", "Epoch 323/400\n", "16/16 - 0s - loss: 0.0293 - accuracy: 1.0000 - val_loss: 0.0358 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 324/400\n", "16/16 - 0s - loss: 0.0292 - accuracy: 1.0000 - val_loss: 0.0357 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 325/400\n", "16/16 - 0s - loss: 0.0290 - accuracy: 1.0000 - val_loss: 0.0355 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 326/400\n", "16/16 - 0s - loss: 0.0289 - accuracy: 1.0000 - val_loss: 0.0354 - val_accuracy: 1.0000 - 125ms/epoch - 8ms/step\n", "Epoch 327/400\n", "16/16 - 0s - loss: 0.0288 - accuracy: 1.0000 - val_loss: 0.0352 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 328/400\n", "16/16 - 0s - loss: 0.0287 - accuracy: 1.0000 - val_loss: 0.0350 - val_accuracy: 1.0000 - 116ms/epoch - 7ms/step\n", "Epoch 329/400\n", "16/16 - 0s - loss: 0.0285 - accuracy: 1.0000 - val_loss: 0.0349 - val_accuracy: 1.0000 - 118ms/epoch - 7ms/step\n", "Epoch 330/400\n", "16/16 - 0s - loss: 0.0284 - accuracy: 1.0000 - val_loss: 0.0348 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 331/400\n", "16/16 - 0s - loss: 0.0283 - accuracy: 1.0000 - val_loss: 0.0346 - val_accuracy: 1.0000 - 76ms/epoch - 5ms/step\n", "Epoch 332/400\n", "16/16 - 0s - loss: 0.0281 - accuracy: 1.0000 - val_loss: 0.0344 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 333/400\n", "16/16 - 0s - loss: 0.0280 - accuracy: 1.0000 - val_loss: 0.0343 - val_accuracy: 1.0000 - 114ms/epoch - 7ms/step\n", "Epoch 334/400\n", "16/16 - 0s - loss: 0.0279 - accuracy: 1.0000 - val_loss: 0.0342 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 335/400\n", "16/16 - 0s - loss: 0.0278 - accuracy: 1.0000 - val_loss: 0.0340 - val_accuracy: 1.0000 - 111ms/epoch - 7ms/step\n", "Epoch 336/400\n", "16/16 - 0s - loss: 0.0277 - accuracy: 1.0000 - val_loss: 0.0338 - val_accuracy: 1.0000 - 113ms/epoch - 7ms/step\n", "Epoch 337/400\n", "16/16 - 0s - loss: 0.0275 - accuracy: 1.0000 - val_loss: 0.0336 - val_accuracy: 1.0000 - 119ms/epoch - 7ms/step\n", "Epoch 338/400\n", "16/16 - 0s - loss: 0.0274 - accuracy: 1.0000 - val_loss: 0.0336 - val_accuracy: 1.0000 - 119ms/epoch - 7ms/step\n", "Epoch 339/400\n", "16/16 - 0s - loss: 0.0273 - accuracy: 1.0000 - val_loss: 0.0334 - val_accuracy: 1.0000 - 113ms/epoch - 7ms/step\n", "Epoch 340/400\n", "16/16 - 0s - loss: 0.0272 - accuracy: 1.0000 - val_loss: 0.0333 - val_accuracy: 1.0000 - 115ms/epoch - 7ms/step\n", "Epoch 341/400\n", "16/16 - 0s - loss: 0.0271 - accuracy: 1.0000 - val_loss: 0.0332 - val_accuracy: 1.0000 - 125ms/epoch - 8ms/step\n", "Epoch 342/400\n", "16/16 - 0s - loss: 0.0270 - accuracy: 1.0000 - val_loss: 0.0331 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 343/400\n", "16/16 - 0s - loss: 0.0269 - accuracy: 1.0000 - val_loss: 0.0328 - val_accuracy: 1.0000 - 113ms/epoch - 7ms/step\n", "Epoch 344/400\n", "16/16 - 0s - loss: 0.0267 - accuracy: 1.0000 - val_loss: 0.0327 - val_accuracy: 1.0000 - 80ms/epoch - 5ms/step\n", "Epoch 345/400\n", "16/16 - 0s - loss: 0.0266 - accuracy: 1.0000 - val_loss: 0.0326 - val_accuracy: 1.0000 - 86ms/epoch - 5ms/step\n", "Epoch 346/400\n", "16/16 - 0s - loss: 0.0265 - accuracy: 1.0000 - val_loss: 0.0324 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 347/400\n", "16/16 - 0s - loss: 0.0264 - accuracy: 1.0000 - val_loss: 0.0323 - val_accuracy: 1.0000 - 114ms/epoch - 7ms/step\n", "Epoch 348/400\n", "16/16 - 0s - loss: 0.0263 - accuracy: 1.0000 - val_loss: 0.0321 - val_accuracy: 1.0000 - 73ms/epoch - 5ms/step\n", "Epoch 349/400\n", "16/16 - 0s - loss: 0.0262 - accuracy: 1.0000 - val_loss: 0.0320 - val_accuracy: 1.0000 - 78ms/epoch - 5ms/step\n", "Epoch 350/400\n", "16/16 - 0s - loss: 0.0261 - accuracy: 1.0000 - val_loss: 0.0319 - val_accuracy: 1.0000 - 80ms/epoch - 5ms/step\n", "Epoch 351/400\n", "16/16 - 0s - loss: 0.0259 - accuracy: 1.0000 - val_loss: 0.0317 - val_accuracy: 1.0000 - 78ms/epoch - 5ms/step\n", "Epoch 352/400\n", "16/16 - 0s - loss: 0.0258 - accuracy: 1.0000 - val_loss: 0.0316 - val_accuracy: 1.0000 - 116ms/epoch - 7ms/step\n", "Epoch 353/400\n", "16/16 - 0s - loss: 0.0257 - accuracy: 1.0000 - val_loss: 0.0315 - val_accuracy: 1.0000 - 112ms/epoch - 7ms/step\n", "Epoch 354/400\n", "16/16 - 0s - loss: 0.0256 - accuracy: 1.0000 - val_loss: 0.0313 - val_accuracy: 1.0000 - 111ms/epoch - 7ms/step\n", "Epoch 355/400\n", "16/16 - 0s - loss: 0.0255 - accuracy: 1.0000 - val_loss: 0.0312 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 356/400\n", "16/16 - 0s - loss: 0.0254 - accuracy: 1.0000 - val_loss: 0.0310 - val_accuracy: 1.0000 - 125ms/epoch - 8ms/step\n", "Epoch 357/400\n", "16/16 - 0s - loss: 0.0253 - accuracy: 1.0000 - val_loss: 0.0309 - val_accuracy: 1.0000 - 76ms/epoch - 5ms/step\n", "Epoch 358/400\n", "16/16 - 0s - loss: 0.0252 - accuracy: 1.0000 - val_loss: 0.0308 - val_accuracy: 1.0000 - 79ms/epoch - 5ms/step\n", "Epoch 359/400\n", "16/16 - 0s - loss: 0.0251 - accuracy: 1.0000 - val_loss: 0.0306 - val_accuracy: 1.0000 - 121ms/epoch - 8ms/step\n", "Epoch 360/400\n", "16/16 - 0s - loss: 0.0250 - accuracy: 1.0000 - val_loss: 0.0305 - val_accuracy: 1.0000 - 112ms/epoch - 7ms/step\n", "Epoch 361/400\n", "16/16 - 0s - loss: 0.0249 - accuracy: 1.0000 - val_loss: 0.0304 - val_accuracy: 1.0000 - 81ms/epoch - 5ms/step\n", "Epoch 362/400\n", "16/16 - 0s - loss: 0.0248 - accuracy: 1.0000 - val_loss: 0.0302 - val_accuracy: 1.0000 - 123ms/epoch - 8ms/step\n", "Epoch 363/400\n", "16/16 - 0s - loss: 0.0247 - accuracy: 1.0000 - val_loss: 0.0301 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 364/400\n", "16/16 - 0s - loss: 0.0246 - accuracy: 1.0000 - val_loss: 0.0299 - val_accuracy: 1.0000 - 119ms/epoch - 7ms/step\n", "Epoch 365/400\n", "16/16 - 0s - loss: 0.0245 - accuracy: 1.0000 - val_loss: 0.0298 - val_accuracy: 1.0000 - 74ms/epoch - 5ms/step\n", "Epoch 366/400\n", "16/16 - 0s - loss: 0.0244 - accuracy: 1.0000 - val_loss: 0.0297 - val_accuracy: 1.0000 - 125ms/epoch - 8ms/step\n", "Epoch 367/400\n", "16/16 - 0s - loss: 0.0243 - accuracy: 1.0000 - val_loss: 0.0296 - val_accuracy: 1.0000 - 76ms/epoch - 5ms/step\n", "Epoch 368/400\n", "16/16 - 0s - loss: 0.0242 - accuracy: 1.0000 - val_loss: 0.0295 - val_accuracy: 1.0000 - 74ms/epoch - 5ms/step\n", "Epoch 369/400\n", "16/16 - 0s - loss: 0.0241 - accuracy: 1.0000 - val_loss: 0.0293 - val_accuracy: 1.0000 - 114ms/epoch - 7ms/step\n", "Epoch 370/400\n", "16/16 - 0s - loss: 0.0240 - accuracy: 1.0000 - val_loss: 0.0292 - val_accuracy: 1.0000 - 78ms/epoch - 5ms/step\n", "Epoch 371/400\n", "16/16 - 0s - loss: 0.0239 - accuracy: 1.0000 - val_loss: 0.0292 - val_accuracy: 1.0000 - 80ms/epoch - 5ms/step\n", "Epoch 372/400\n", "16/16 - 0s - loss: 0.0238 - accuracy: 1.0000 - val_loss: 0.0290 - val_accuracy: 1.0000 - 113ms/epoch - 7ms/step\n", "Epoch 373/400\n", "16/16 - 0s - loss: 0.0237 - accuracy: 1.0000 - val_loss: 0.0289 - val_accuracy: 1.0000 - 121ms/epoch - 8ms/step\n", "Epoch 374/400\n", "16/16 - 0s - loss: 0.0236 - accuracy: 1.0000 - val_loss: 0.0288 - val_accuracy: 1.0000 - 113ms/epoch - 7ms/step\n", "Epoch 375/400\n", "16/16 - 0s - loss: 0.0235 - accuracy: 1.0000 - val_loss: 0.0287 - val_accuracy: 1.0000 - 77ms/epoch - 5ms/step\n", "Epoch 376/400\n", "16/16 - 0s - loss: 0.0234 - accuracy: 1.0000 - val_loss: 0.0286 - val_accuracy: 1.0000 - 84ms/epoch - 5ms/step\n", "Epoch 377/400\n", "16/16 - 0s - loss: 0.0233 - accuracy: 1.0000 - val_loss: 0.0284 - val_accuracy: 1.0000 - 80ms/epoch - 5ms/step\n", "Epoch 378/400\n", "16/16 - 0s - loss: 0.0232 - accuracy: 1.0000 - val_loss: 0.0283 - val_accuracy: 1.0000 - 117ms/epoch - 7ms/step\n", "Epoch 379/400\n", "16/16 - 0s - loss: 0.0231 - accuracy: 1.0000 - val_loss: 0.0282 - val_accuracy: 1.0000 - 114ms/epoch - 7ms/step\n", "Epoch 380/400\n", "16/16 - 0s - loss: 0.0230 - accuracy: 1.0000 - val_loss: 0.0281 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 381/400\n", "16/16 - 0s - loss: 0.0229 - accuracy: 1.0000 - val_loss: 0.0280 - val_accuracy: 1.0000 - 116ms/epoch - 7ms/step\n", "Epoch 382/400\n", "16/16 - 0s - loss: 0.0228 - accuracy: 1.0000 - val_loss: 0.0279 - val_accuracy: 1.0000 - 78ms/epoch - 5ms/step\n", "Epoch 383/400\n", "16/16 - 0s - loss: 0.0227 - accuracy: 1.0000 - val_loss: 0.0278 - val_accuracy: 1.0000 - 72ms/epoch - 4ms/step\n", "Epoch 384/400\n", "16/16 - 0s - loss: 0.0226 - accuracy: 1.0000 - val_loss: 0.0276 - val_accuracy: 1.0000 - 124ms/epoch - 8ms/step\n", "Epoch 385/400\n", "16/16 - 0s - loss: 0.0225 - accuracy: 1.0000 - val_loss: 0.0276 - val_accuracy: 1.0000 - 114ms/epoch - 7ms/step\n", "Epoch 386/400\n", "16/16 - 0s - loss: 0.0224 - accuracy: 1.0000 - val_loss: 0.0274 - val_accuracy: 1.0000 - 84ms/epoch - 5ms/step\n", "Epoch 387/400\n", "16/16 - 0s - loss: 0.0223 - accuracy: 1.0000 - val_loss: 0.0273 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 388/400\n", "16/16 - 0s - loss: 0.0223 - accuracy: 1.0000 - val_loss: 0.0272 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 389/400\n", "16/16 - 0s - loss: 0.0222 - accuracy: 1.0000 - val_loss: 0.0271 - val_accuracy: 1.0000 - 85ms/epoch - 5ms/step\n", "Epoch 390/400\n", "16/16 - 0s - loss: 0.0221 - accuracy: 1.0000 - val_loss: 0.0270 - val_accuracy: 1.0000 - 77ms/epoch - 5ms/step\n", "Epoch 391/400\n", "16/16 - 0s - loss: 0.0220 - accuracy: 1.0000 - val_loss: 0.0269 - val_accuracy: 1.0000 - 118ms/epoch - 7ms/step\n", "Epoch 392/400\n", "16/16 - 0s - loss: 0.0219 - accuracy: 1.0000 - val_loss: 0.0268 - val_accuracy: 1.0000 - 117ms/epoch - 7ms/step\n", "Epoch 393/400\n", "16/16 - 0s - loss: 0.0218 - accuracy: 1.0000 - val_loss: 0.0266 - val_accuracy: 1.0000 - 71ms/epoch - 4ms/step\n", "Epoch 394/400\n", "16/16 - 0s - loss: 0.0217 - accuracy: 1.0000 - val_loss: 0.0266 - val_accuracy: 1.0000 - 117ms/epoch - 7ms/step\n", "Epoch 395/400\n", "16/16 - 0s - loss: 0.0217 - accuracy: 1.0000 - val_loss: 0.0264 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 396/400\n", "16/16 - 0s - loss: 0.0216 - accuracy: 1.0000 - val_loss: 0.0264 - val_accuracy: 1.0000 - 77ms/epoch - 5ms/step\n", "Epoch 397/400\n", "16/16 - 0s - loss: 0.0215 - accuracy: 1.0000 - val_loss: 0.0263 - val_accuracy: 1.0000 - 124ms/epoch - 8ms/step\n", "Epoch 398/400\n", "16/16 - 0s - loss: 0.0214 - accuracy: 1.0000 - val_loss: 0.0262 - val_accuracy: 1.0000 - 118ms/epoch - 7ms/step\n", "Epoch 399/400\n", "16/16 - 0s - loss: 0.0213 - accuracy: 1.0000 - val_loss: 0.0261 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 400/400\n", "16/16 - 0s - loss: 0.0212 - accuracy: 1.0000 - val_loss: 0.0260 - val_accuracy: 1.0000 - 110ms/epoch - 7ms/step\n", "第4个弱分类器训练完毕\n", "Epoch 1/400\n", "16/16 - 1s - loss: 1.9987 - accuracy: 0.5900 - val_loss: 2.1152 - val_accuracy: 0.6080 - 964ms/epoch - 60ms/step\n", "Epoch 2/400\n", "16/16 - 0s - loss: 1.9738 - accuracy: 0.5900 - val_loss: 2.0892 - val_accuracy: 0.6080 - 111ms/epoch - 7ms/step\n", "Epoch 3/400\n", "16/16 - 0s - loss: 1.9491 - accuracy: 0.5920 - val_loss: 2.0631 - val_accuracy: 0.6100 - 115ms/epoch - 7ms/step\n", "Epoch 4/400\n", "16/16 - 0s - loss: 1.9249 - accuracy: 0.5940 - val_loss: 2.0369 - val_accuracy: 0.6100 - 113ms/epoch - 7ms/step\n", "Epoch 5/400\n", "16/16 - 0s - loss: 1.9007 - accuracy: 0.5960 - val_loss: 2.0114 - val_accuracy: 0.6140 - 79ms/epoch - 5ms/step\n", "Epoch 6/400\n", "16/16 - 0s - loss: 1.8767 - accuracy: 0.5980 - val_loss: 1.9863 - val_accuracy: 0.6160 - 76ms/epoch - 5ms/step\n", "Epoch 7/400\n", "16/16 - 0s - loss: 1.8530 - accuracy: 0.6020 - val_loss: 1.9614 - val_accuracy: 0.6160 - 86ms/epoch - 5ms/step\n", "Epoch 8/400\n", "16/16 - 0s - loss: 1.8294 - accuracy: 0.6040 - val_loss: 1.9370 - val_accuracy: 0.6180 - 112ms/epoch - 7ms/step\n", "Epoch 9/400\n", "16/16 - 0s - loss: 1.8064 - accuracy: 0.6040 - val_loss: 1.9120 - val_accuracy: 0.6200 - 119ms/epoch - 7ms/step\n", "Epoch 10/400\n", "16/16 - 0s - loss: 1.7831 - accuracy: 0.6080 - val_loss: 1.8876 - val_accuracy: 0.6220 - 74ms/epoch - 5ms/step\n", "Epoch 11/400\n", "16/16 - 0s - loss: 1.7600 - accuracy: 0.6100 - val_loss: 1.8637 - val_accuracy: 0.6220 - 72ms/epoch - 4ms/step\n", "Epoch 12/400\n", "16/16 - 0s - loss: 1.7379 - accuracy: 0.6120 - val_loss: 1.8387 - val_accuracy: 0.6240 - 77ms/epoch - 5ms/step\n", "Epoch 13/400\n", "16/16 - 0s - loss: 1.7143 - accuracy: 0.6160 - val_loss: 1.8153 - val_accuracy: 0.6240 - 122ms/epoch - 8ms/step\n", "Epoch 14/400\n", "16/16 - 0s - loss: 1.6920 - accuracy: 0.6160 - val_loss: 1.7913 - val_accuracy: 0.6240 - 74ms/epoch - 5ms/step\n", "Epoch 15/400\n", "16/16 - 0s - loss: 1.6700 - accuracy: 0.6180 - val_loss: 1.7672 - val_accuracy: 0.6240 - 73ms/epoch - 5ms/step\n", "Epoch 16/400\n", "16/16 - 0s - loss: 1.6472 - accuracy: 0.6180 - val_loss: 1.7445 - val_accuracy: 0.6240 - 79ms/epoch - 5ms/step\n", "Epoch 17/400\n", "16/16 - 0s - loss: 1.6264 - accuracy: 0.6200 - val_loss: 1.7203 - val_accuracy: 0.6260 - 113ms/epoch - 7ms/step\n", "Epoch 18/400\n", "16/16 - 0s - loss: 1.6041 - accuracy: 0.6200 - val_loss: 1.6976 - val_accuracy: 0.6260 - 75ms/epoch - 5ms/step\n", "Epoch 19/400\n", "16/16 - 0s - loss: 1.5833 - accuracy: 0.6260 - val_loss: 1.6742 - val_accuracy: 0.6280 - 72ms/epoch - 4ms/step\n", "Epoch 20/400\n", "16/16 - 0s - loss: 1.5618 - accuracy: 0.6280 - val_loss: 1.6519 - val_accuracy: 0.6300 - 73ms/epoch - 5ms/step\n", "Epoch 21/400\n", "16/16 - 0s - loss: 1.5411 - accuracy: 0.6280 - val_loss: 1.6295 - val_accuracy: 0.6320 - 110ms/epoch - 7ms/step\n", "Epoch 22/400\n", "16/16 - 0s - loss: 1.5201 - accuracy: 0.6340 - val_loss: 1.6077 - val_accuracy: 0.6340 - 117ms/epoch - 7ms/step\n", "Epoch 23/400\n", "16/16 - 0s - loss: 1.4997 - accuracy: 0.6360 - val_loss: 1.5860 - val_accuracy: 0.6340 - 72ms/epoch - 5ms/step\n", "Epoch 24/400\n", "16/16 - 0s - loss: 1.4796 - accuracy: 0.6380 - val_loss: 1.5634 - val_accuracy: 0.6360 - 75ms/epoch - 5ms/step\n", "Epoch 25/400\n", "16/16 - 0s - loss: 1.4592 - accuracy: 0.6440 - val_loss: 1.5417 - val_accuracy: 0.6420 - 77ms/epoch - 5ms/step\n", "Epoch 26/400\n", "16/16 - 0s - loss: 1.4392 - accuracy: 0.6440 - val_loss: 1.5199 - val_accuracy: 0.6420 - 73ms/epoch - 5ms/step\n", "Epoch 27/400\n", "16/16 - 0s - loss: 1.4189 - accuracy: 0.6480 - val_loss: 1.4987 - val_accuracy: 0.6440 - 82ms/epoch - 5ms/step\n", "Epoch 28/400\n", "16/16 - 0s - loss: 1.3993 - accuracy: 0.6520 - val_loss: 1.4773 - val_accuracy: 0.6460 - 77ms/epoch - 5ms/step\n", "Epoch 29/400\n", "16/16 - 0s - loss: 1.3794 - accuracy: 0.6520 - val_loss: 1.4567 - val_accuracy: 0.6460 - 80ms/epoch - 5ms/step\n", "Epoch 30/400\n", "16/16 - 0s - loss: 1.3603 - accuracy: 0.6540 - val_loss: 1.4353 - val_accuracy: 0.6520 - 77ms/epoch - 5ms/step\n", "Epoch 31/400\n", "16/16 - 0s - loss: 1.3407 - accuracy: 0.6540 - val_loss: 1.4143 - val_accuracy: 0.6540 - 73ms/epoch - 5ms/step\n", "Epoch 32/400\n", "16/16 - 0s - loss: 1.3217 - accuracy: 0.6560 - val_loss: 1.3935 - val_accuracy: 0.6540 - 112ms/epoch - 7ms/step\n", "Epoch 33/400\n", "16/16 - 0s - loss: 1.3029 - accuracy: 0.6560 - val_loss: 1.3728 - val_accuracy: 0.6540 - 116ms/epoch - 7ms/step\n", "Epoch 34/400\n", "16/16 - 0s - loss: 1.2841 - accuracy: 0.6560 - val_loss: 1.3529 - val_accuracy: 0.6540 - 117ms/epoch - 7ms/step\n", "Epoch 35/400\n", "16/16 - 0s - loss: 1.2657 - accuracy: 0.6580 - val_loss: 1.3334 - val_accuracy: 0.6540 - 117ms/epoch - 7ms/step\n", "Epoch 36/400\n", "16/16 - 0s - loss: 1.2477 - accuracy: 0.6580 - val_loss: 1.3137 - val_accuracy: 0.6560 - 115ms/epoch - 7ms/step\n", "Epoch 37/400\n", "16/16 - 0s - loss: 1.2296 - accuracy: 0.6600 - val_loss: 1.2943 - val_accuracy: 0.6560 - 76ms/epoch - 5ms/step\n", "Epoch 38/400\n", "16/16 - 0s - loss: 1.2124 - accuracy: 0.6620 - val_loss: 1.2743 - val_accuracy: 0.6580 - 77ms/epoch - 5ms/step\n", "Epoch 39/400\n", "16/16 - 0s - loss: 1.1938 - accuracy: 0.6640 - val_loss: 1.2559 - val_accuracy: 0.6660 - 84ms/epoch - 5ms/step\n", "Epoch 40/400\n", "16/16 - 0s - loss: 1.1764 - accuracy: 0.6640 - val_loss: 1.2373 - val_accuracy: 0.6660 - 111ms/epoch - 7ms/step\n", "Epoch 41/400\n", "16/16 - 0s - loss: 1.1594 - accuracy: 0.6640 - val_loss: 1.2180 - val_accuracy: 0.6660 - 72ms/epoch - 5ms/step\n", "Epoch 42/400\n", "16/16 - 0s - loss: 1.1421 - accuracy: 0.6640 - val_loss: 1.1992 - val_accuracy: 0.6660 - 73ms/epoch - 5ms/step\n", "Epoch 43/400\n", "16/16 - 0s - loss: 1.1246 - accuracy: 0.6660 - val_loss: 1.1815 - val_accuracy: 0.6660 - 74ms/epoch - 5ms/step\n", "Epoch 44/400\n", "16/16 - 0s - loss: 1.1085 - accuracy: 0.6660 - val_loss: 1.1628 - val_accuracy: 0.6660 - 77ms/epoch - 5ms/step\n", "Epoch 45/400\n", "16/16 - 0s - loss: 1.0912 - accuracy: 0.6660 - val_loss: 1.1456 - val_accuracy: 0.6660 - 118ms/epoch - 7ms/step\n", "Epoch 46/400\n", "16/16 - 0s - loss: 1.0753 - accuracy: 0.6700 - val_loss: 1.1278 - val_accuracy: 0.6660 - 77ms/epoch - 5ms/step\n", "Epoch 47/400\n", "16/16 - 0s - loss: 1.0591 - accuracy: 0.6740 - val_loss: 1.1103 - val_accuracy: 0.6680 - 70ms/epoch - 4ms/step\n", "Epoch 48/400\n", "16/16 - 0s - loss: 1.0433 - accuracy: 0.6780 - val_loss: 1.0929 - val_accuracy: 0.6700 - 81ms/epoch - 5ms/step\n", "Epoch 49/400\n", "16/16 - 0s - loss: 1.0273 - accuracy: 0.6780 - val_loss: 1.0758 - val_accuracy: 0.6720 - 76ms/epoch - 5ms/step\n", "Epoch 50/400\n", "16/16 - 0s - loss: 1.0117 - accuracy: 0.6780 - val_loss: 1.0589 - val_accuracy: 0.6760 - 116ms/epoch - 7ms/step\n", "Epoch 51/400\n", "16/16 - 0s - loss: 0.9963 - accuracy: 0.6800 - val_loss: 1.0422 - val_accuracy: 0.6780 - 83ms/epoch - 5ms/step\n", "Epoch 52/400\n", "16/16 - 0s - loss: 0.9810 - accuracy: 0.6840 - val_loss: 1.0259 - val_accuracy: 0.6800 - 78ms/epoch - 5ms/step\n", "Epoch 53/400\n", "16/16 - 0s - loss: 0.9661 - accuracy: 0.6860 - val_loss: 1.0100 - val_accuracy: 0.6800 - 78ms/epoch - 5ms/step\n", "Epoch 54/400\n", "16/16 - 0s - loss: 0.9511 - accuracy: 0.6880 - val_loss: 0.9943 - val_accuracy: 0.6820 - 118ms/epoch - 7ms/step\n", "Epoch 55/400\n", "16/16 - 0s - loss: 0.9367 - accuracy: 0.6880 - val_loss: 0.9783 - val_accuracy: 0.6880 - 75ms/epoch - 5ms/step\n", "Epoch 56/400\n", "16/16 - 0s - loss: 0.9222 - accuracy: 0.6920 - val_loss: 0.9626 - val_accuracy: 0.6900 - 123ms/epoch - 8ms/step\n", "Epoch 57/400\n", "16/16 - 0s - loss: 0.9077 - accuracy: 0.6960 - val_loss: 0.9475 - val_accuracy: 0.6940 - 78ms/epoch - 5ms/step\n", "Epoch 58/400\n", "16/16 - 0s - loss: 0.8941 - accuracy: 0.7000 - val_loss: 0.9321 - val_accuracy: 0.6980 - 74ms/epoch - 5ms/step\n", "Epoch 59/400\n", "16/16 - 0s - loss: 0.8799 - accuracy: 0.7000 - val_loss: 0.9175 - val_accuracy: 0.7000 - 122ms/epoch - 8ms/step\n", "Epoch 60/400\n", "16/16 - 0s - loss: 0.8663 - accuracy: 0.7020 - val_loss: 0.9029 - val_accuracy: 0.7000 - 149ms/epoch - 9ms/step\n", "Epoch 61/400\n", "16/16 - 0s - loss: 0.8528 - accuracy: 0.7020 - val_loss: 0.8886 - val_accuracy: 0.7000 - 143ms/epoch - 9ms/step\n", "Epoch 62/400\n", "16/16 - 0s - loss: 0.8397 - accuracy: 0.7020 - val_loss: 0.8742 - val_accuracy: 0.7000 - 143ms/epoch - 9ms/step\n", "Epoch 63/400\n", "16/16 - 0s - loss: 0.8263 - accuracy: 0.7020 - val_loss: 0.8602 - val_accuracy: 0.7040 - 125ms/epoch - 8ms/step\n", "Epoch 64/400\n", "16/16 - 0s - loss: 0.8135 - accuracy: 0.7040 - val_loss: 0.8465 - val_accuracy: 0.7040 - 108ms/epoch - 7ms/step\n", "Epoch 65/400\n", "16/16 - 0s - loss: 0.8007 - accuracy: 0.7100 - val_loss: 0.8326 - val_accuracy: 0.7060 - 146ms/epoch - 9ms/step\n", "Epoch 66/400\n", "16/16 - 0s - loss: 0.7879 - accuracy: 0.7120 - val_loss: 0.8193 - val_accuracy: 0.7120 - 127ms/epoch - 8ms/step\n", "Epoch 67/400\n", "16/16 - 0s - loss: 0.7755 - accuracy: 0.7120 - val_loss: 0.8059 - val_accuracy: 0.7160 - 111ms/epoch - 7ms/step\n", "Epoch 68/400\n", "16/16 - 0s - loss: 0.7634 - accuracy: 0.7160 - val_loss: 0.7926 - val_accuracy: 0.7220 - 172ms/epoch - 11ms/step\n", "Epoch 69/400\n", "16/16 - 0s - loss: 0.7511 - accuracy: 0.7160 - val_loss: 0.7800 - val_accuracy: 0.7260 - 164ms/epoch - 10ms/step\n", "Epoch 70/400\n", "16/16 - 0s - loss: 0.7395 - accuracy: 0.7160 - val_loss: 0.7674 - val_accuracy: 0.7280 - 101ms/epoch - 6ms/step\n", "Epoch 71/400\n", "16/16 - 0s - loss: 0.7276 - accuracy: 0.7160 - val_loss: 0.7554 - val_accuracy: 0.7320 - 104ms/epoch - 6ms/step\n", "Epoch 72/400\n", "16/16 - 0s - loss: 0.7163 - accuracy: 0.7200 - val_loss: 0.7431 - val_accuracy: 0.7360 - 102ms/epoch - 6ms/step\n", "Epoch 73/400\n", "16/16 - 0s - loss: 0.7050 - accuracy: 0.7200 - val_loss: 0.7311 - val_accuracy: 0.7360 - 113ms/epoch - 7ms/step\n", "Epoch 74/400\n", "16/16 - 0s - loss: 0.6939 - accuracy: 0.7240 - val_loss: 0.7189 - val_accuracy: 0.7380 - 105ms/epoch - 7ms/step\n", "Epoch 75/400\n", "16/16 - 0s - loss: 0.6830 - accuracy: 0.7240 - val_loss: 0.7069 - val_accuracy: 0.7420 - 96ms/epoch - 6ms/step\n", "Epoch 76/400\n", "16/16 - 0s - loss: 0.6718 - accuracy: 0.7340 - val_loss: 0.6959 - val_accuracy: 0.7460 - 134ms/epoch - 8ms/step\n", "Epoch 77/400\n", "16/16 - 0s - loss: 0.6615 - accuracy: 0.7380 - val_loss: 0.6844 - val_accuracy: 0.7460 - 143ms/epoch - 9ms/step\n", "Epoch 78/400\n", "16/16 - 0s - loss: 0.6514 - accuracy: 0.7420 - val_loss: 0.6731 - val_accuracy: 0.7480 - 116ms/epoch - 7ms/step\n", "Epoch 79/400\n", "16/16 - 0s - loss: 0.6407 - accuracy: 0.7460 - val_loss: 0.6624 - val_accuracy: 0.7520 - 103ms/epoch - 6ms/step\n", "Epoch 80/400\n", "16/16 - 0s - loss: 0.6306 - accuracy: 0.7600 - val_loss: 0.6522 - val_accuracy: 0.7580 - 141ms/epoch - 9ms/step\n", "Epoch 81/400\n", "16/16 - 0s - loss: 0.6209 - accuracy: 0.7660 - val_loss: 0.6418 - val_accuracy: 0.7600 - 86ms/epoch - 5ms/step\n", "Epoch 82/400\n", "16/16 - 0s - loss: 0.6114 - accuracy: 0.7680 - val_loss: 0.6316 - val_accuracy: 0.7620 - 115ms/epoch - 7ms/step\n", "Epoch 83/400\n", "16/16 - 0s - loss: 0.6019 - accuracy: 0.7740 - val_loss: 0.6217 - val_accuracy: 0.7660 - 115ms/epoch - 7ms/step\n", "Epoch 84/400\n", "16/16 - 0s - loss: 0.5925 - accuracy: 0.7820 - val_loss: 0.6121 - val_accuracy: 0.7660 - 116ms/epoch - 7ms/step\n", "Epoch 85/400\n", "16/16 - 0s - loss: 0.5835 - accuracy: 0.7840 - val_loss: 0.6026 - val_accuracy: 0.7680 - 83ms/epoch - 5ms/step\n", "Epoch 86/400\n", "16/16 - 0s - loss: 0.5747 - accuracy: 0.7860 - val_loss: 0.5930 - val_accuracy: 0.7700 - 117ms/epoch - 7ms/step\n", "Epoch 87/400\n", "16/16 - 0s - loss: 0.5659 - accuracy: 0.7920 - val_loss: 0.5835 - val_accuracy: 0.7700 - 72ms/epoch - 5ms/step\n", "Epoch 88/400\n", "16/16 - 0s - loss: 0.5574 - accuracy: 0.7980 - val_loss: 0.5744 - val_accuracy: 0.7740 - 109ms/epoch - 7ms/step\n", "Epoch 89/400\n", "16/16 - 0s - loss: 0.5488 - accuracy: 0.8020 - val_loss: 0.5661 - val_accuracy: 0.7820 - 114ms/epoch - 7ms/step\n", "Epoch 90/400\n", "16/16 - 0s - loss: 0.5407 - accuracy: 0.8020 - val_loss: 0.5580 - val_accuracy: 0.7820 - 74ms/epoch - 5ms/step\n", "Epoch 91/400\n", "16/16 - 0s - loss: 0.5330 - accuracy: 0.8020 - val_loss: 0.5492 - val_accuracy: 0.7840 - 120ms/epoch - 7ms/step\n", "Epoch 92/400\n", "16/16 - 0s - loss: 0.5251 - accuracy: 0.8020 - val_loss: 0.5408 - val_accuracy: 0.7880 - 75ms/epoch - 5ms/step\n", "Epoch 93/400\n", "16/16 - 0s - loss: 0.5173 - accuracy: 0.8100 - val_loss: 0.5332 - val_accuracy: 0.7880 - 73ms/epoch - 5ms/step\n", "Epoch 94/400\n", "16/16 - 0s - loss: 0.5101 - accuracy: 0.8080 - val_loss: 0.5252 - val_accuracy: 0.7900 - 71ms/epoch - 4ms/step\n", "Epoch 95/400\n", "16/16 - 0s - loss: 0.5028 - accuracy: 0.8080 - val_loss: 0.5174 - val_accuracy: 0.7940 - 116ms/epoch - 7ms/step\n", "Epoch 96/400\n", "16/16 - 0s - loss: 0.4956 - accuracy: 0.8120 - val_loss: 0.5103 - val_accuracy: 0.7960 - 117ms/epoch - 7ms/step\n", "Epoch 97/400\n", "16/16 - 0s - loss: 0.4888 - accuracy: 0.8160 - val_loss: 0.5033 - val_accuracy: 0.7960 - 114ms/epoch - 7ms/step\n", "Epoch 98/400\n", "16/16 - 0s - loss: 0.4823 - accuracy: 0.8180 - val_loss: 0.4964 - val_accuracy: 0.8000 - 80ms/epoch - 5ms/step\n", "Epoch 99/400\n", "16/16 - 0s - loss: 0.4757 - accuracy: 0.8180 - val_loss: 0.4899 - val_accuracy: 0.8040 - 116ms/epoch - 7ms/step\n", "Epoch 100/400\n", "16/16 - 0s - loss: 0.4699 - accuracy: 0.8200 - val_loss: 0.4826 - val_accuracy: 0.8040 - 83ms/epoch - 5ms/step\n", "Epoch 101/400\n", "16/16 - 0s - loss: 0.4630 - accuracy: 0.8240 - val_loss: 0.4770 - val_accuracy: 0.8080 - 119ms/epoch - 7ms/step\n", "Epoch 102/400\n", "16/16 - 0s - loss: 0.4575 - accuracy: 0.8280 - val_loss: 0.4707 - val_accuracy: 0.8080 - 73ms/epoch - 5ms/step\n", "Epoch 103/400\n", "16/16 - 0s - loss: 0.4518 - accuracy: 0.8300 - val_loss: 0.4644 - val_accuracy: 0.8140 - 110ms/epoch - 7ms/step\n", "Epoch 104/400\n", "16/16 - 0s - loss: 0.4459 - accuracy: 0.8300 - val_loss: 0.4589 - val_accuracy: 0.8140 - 84ms/epoch - 5ms/step\n", "Epoch 105/400\n", "16/16 - 0s - loss: 0.4408 - accuracy: 0.8300 - val_loss: 0.4532 - val_accuracy: 0.8140 - 124ms/epoch - 8ms/step\n", "Epoch 106/400\n", "16/16 - 0s - loss: 0.4354 - accuracy: 0.8340 - val_loss: 0.4482 - val_accuracy: 0.8180 - 117ms/epoch - 7ms/step\n", "Epoch 107/400\n", "16/16 - 0s - loss: 0.4305 - accuracy: 0.8360 - val_loss: 0.4431 - val_accuracy: 0.8220 - 72ms/epoch - 5ms/step\n", "Epoch 108/400\n", "16/16 - 0s - loss: 0.4255 - accuracy: 0.8360 - val_loss: 0.4383 - val_accuracy: 0.8220 - 76ms/epoch - 5ms/step\n", "Epoch 109/400\n", "16/16 - 0s - loss: 0.4209 - accuracy: 0.8400 - val_loss: 0.4334 - val_accuracy: 0.8240 - 111ms/epoch - 7ms/step\n", "Epoch 110/400\n", "16/16 - 0s - loss: 0.4164 - accuracy: 0.8400 - val_loss: 0.4288 - val_accuracy: 0.8260 - 76ms/epoch - 5ms/step\n", "Epoch 111/400\n", "16/16 - 0s - loss: 0.4120 - accuracy: 0.8400 - val_loss: 0.4241 - val_accuracy: 0.8280 - 116ms/epoch - 7ms/step\n", "Epoch 112/400\n", "16/16 - 0s - loss: 0.4077 - accuracy: 0.8420 - val_loss: 0.4197 - val_accuracy: 0.8300 - 115ms/epoch - 7ms/step\n", "Epoch 113/400\n", "16/16 - 0s - loss: 0.4035 - accuracy: 0.8420 - val_loss: 0.4157 - val_accuracy: 0.8320 - 116ms/epoch - 7ms/step\n", "Epoch 114/400\n", "16/16 - 0s - loss: 0.3998 - accuracy: 0.8420 - val_loss: 0.4115 - val_accuracy: 0.8340 - 116ms/epoch - 7ms/step\n", "Epoch 115/400\n", "16/16 - 0s - loss: 0.3958 - accuracy: 0.8420 - val_loss: 0.4076 - val_accuracy: 0.8360 - 81ms/epoch - 5ms/step\n", "Epoch 116/400\n", "16/16 - 0s - loss: 0.3920 - accuracy: 0.8460 - val_loss: 0.4040 - val_accuracy: 0.8400 - 79ms/epoch - 5ms/step\n", "Epoch 117/400\n", "16/16 - 0s - loss: 0.3885 - accuracy: 0.8480 - val_loss: 0.4004 - val_accuracy: 0.8400 - 116ms/epoch - 7ms/step\n", "Epoch 118/400\n", "16/16 - 0s - loss: 0.3850 - accuracy: 0.8480 - val_loss: 0.3970 - val_accuracy: 0.8400 - 81ms/epoch - 5ms/step\n", "Epoch 119/400\n", "16/16 - 0s - loss: 0.3818 - accuracy: 0.8480 - val_loss: 0.3936 - val_accuracy: 0.8420 - 72ms/epoch - 4ms/step\n", "Epoch 120/400\n", "16/16 - 0s - loss: 0.3785 - accuracy: 0.8480 - val_loss: 0.3905 - val_accuracy: 0.8440 - 116ms/epoch - 7ms/step\n", "Epoch 121/400\n", "16/16 - 0s - loss: 0.3755 - accuracy: 0.8480 - val_loss: 0.3873 - val_accuracy: 0.8440 - 70ms/epoch - 4ms/step\n", "Epoch 122/400\n", "16/16 - 0s - loss: 0.3725 - accuracy: 0.8480 - val_loss: 0.3845 - val_accuracy: 0.8460 - 74ms/epoch - 5ms/step\n", "Epoch 123/400\n", "16/16 - 0s - loss: 0.3697 - accuracy: 0.8520 - val_loss: 0.3816 - val_accuracy: 0.8500 - 116ms/epoch - 7ms/step\n", "Epoch 124/400\n", "16/16 - 0s - loss: 0.3670 - accuracy: 0.8520 - val_loss: 0.3788 - val_accuracy: 0.8500 - 78ms/epoch - 5ms/step\n", "Epoch 125/400\n", "16/16 - 0s - loss: 0.3644 - accuracy: 0.8540 - val_loss: 0.3762 - val_accuracy: 0.8520 - 83ms/epoch - 5ms/step\n", "Epoch 126/400\n", "16/16 - 0s - loss: 0.3617 - accuracy: 0.8540 - val_loss: 0.3737 - val_accuracy: 0.8540 - 120ms/epoch - 7ms/step\n", "Epoch 127/400\n", "16/16 - 0s - loss: 0.3592 - accuracy: 0.8540 - val_loss: 0.3710 - val_accuracy: 0.8560 - 84ms/epoch - 5ms/step\n", "Epoch 128/400\n", "16/16 - 0s - loss: 0.3567 - accuracy: 0.8520 - val_loss: 0.3687 - val_accuracy: 0.8580 - 119ms/epoch - 7ms/step\n", "Epoch 129/400\n", "16/16 - 0s - loss: 0.3544 - accuracy: 0.8520 - val_loss: 0.3662 - val_accuracy: 0.8580 - 78ms/epoch - 5ms/step\n", "Epoch 130/400\n", "16/16 - 0s - loss: 0.3521 - accuracy: 0.8540 - val_loss: 0.3638 - val_accuracy: 0.8600 - 117ms/epoch - 7ms/step\n", "Epoch 131/400\n", "16/16 - 0s - loss: 0.3499 - accuracy: 0.8640 - val_loss: 0.3615 - val_accuracy: 0.8600 - 114ms/epoch - 7ms/step\n", "Epoch 132/400\n", "16/16 - 0s - loss: 0.3477 - accuracy: 0.8640 - val_loss: 0.3594 - val_accuracy: 0.8640 - 74ms/epoch - 5ms/step\n", "Epoch 133/400\n", "16/16 - 0s - loss: 0.3456 - accuracy: 0.8680 - val_loss: 0.3573 - val_accuracy: 0.8700 - 114ms/epoch - 7ms/step\n", "Epoch 134/400\n", "16/16 - 0s - loss: 0.3436 - accuracy: 0.8680 - val_loss: 0.3552 - val_accuracy: 0.8720 - 113ms/epoch - 7ms/step\n", "Epoch 135/400\n", "16/16 - 0s - loss: 0.3415 - accuracy: 0.8700 - val_loss: 0.3533 - val_accuracy: 0.8700 - 76ms/epoch - 5ms/step\n", "Epoch 136/400\n", "16/16 - 0s - loss: 0.3397 - accuracy: 0.8700 - val_loss: 0.3513 - val_accuracy: 0.8720 - 127ms/epoch - 8ms/step\n", "Epoch 137/400\n", "16/16 - 0s - loss: 0.3377 - accuracy: 0.8700 - val_loss: 0.3494 - val_accuracy: 0.8720 - 119ms/epoch - 7ms/step\n", "Epoch 138/400\n", "16/16 - 0s - loss: 0.3359 - accuracy: 0.8720 - val_loss: 0.3475 - val_accuracy: 0.8740 - 74ms/epoch - 5ms/step\n", "Epoch 139/400\n", "16/16 - 0s - loss: 0.3340 - accuracy: 0.8720 - val_loss: 0.3456 - val_accuracy: 0.8780 - 77ms/epoch - 5ms/step\n", "Epoch 140/400\n", "16/16 - 0s - loss: 0.3322 - accuracy: 0.8740 - val_loss: 0.3439 - val_accuracy: 0.8800 - 121ms/epoch - 8ms/step\n", "Epoch 141/400\n", "16/16 - 0s - loss: 0.3305 - accuracy: 0.8760 - val_loss: 0.3420 - val_accuracy: 0.8880 - 76ms/epoch - 5ms/step\n", "Epoch 142/400\n", "16/16 - 0s - loss: 0.3287 - accuracy: 0.8780 - val_loss: 0.3404 - val_accuracy: 0.8900 - 116ms/epoch - 7ms/step\n", "Epoch 143/400\n", "16/16 - 0s - loss: 0.3271 - accuracy: 0.8860 - val_loss: 0.3385 - val_accuracy: 0.8940 - 119ms/epoch - 7ms/step\n", "Epoch 144/400\n", "16/16 - 0s - loss: 0.3252 - accuracy: 0.8900 - val_loss: 0.3369 - val_accuracy: 0.8960 - 90ms/epoch - 6ms/step\n", "Epoch 145/400\n", "16/16 - 0s - loss: 0.3236 - accuracy: 0.8920 - val_loss: 0.3352 - val_accuracy: 0.8980 - 111ms/epoch - 7ms/step\n", "Epoch 146/400\n", "16/16 - 0s - loss: 0.3220 - accuracy: 0.8980 - val_loss: 0.3334 - val_accuracy: 0.9000 - 72ms/epoch - 5ms/step\n", "Epoch 147/400\n", "16/16 - 0s - loss: 0.3203 - accuracy: 0.9000 - val_loss: 0.3318 - val_accuracy: 0.9020 - 82ms/epoch - 5ms/step\n", "Epoch 148/400\n", "16/16 - 0s - loss: 0.3188 - accuracy: 0.9000 - val_loss: 0.3301 - val_accuracy: 0.9060 - 119ms/epoch - 7ms/step\n", "Epoch 149/400\n", "16/16 - 0s - loss: 0.3172 - accuracy: 0.9060 - val_loss: 0.3286 - val_accuracy: 0.9080 - 75ms/epoch - 5ms/step\n", "Epoch 150/400\n", "16/16 - 0s - loss: 0.3156 - accuracy: 0.9100 - val_loss: 0.3270 - val_accuracy: 0.9100 - 114ms/epoch - 7ms/step\n", "Epoch 151/400\n", "16/16 - 0s - loss: 0.3141 - accuracy: 0.9100 - val_loss: 0.3254 - val_accuracy: 0.9140 - 115ms/epoch - 7ms/step\n", "Epoch 152/400\n", "16/16 - 0s - loss: 0.3125 - accuracy: 0.9100 - val_loss: 0.3238 - val_accuracy: 0.9140 - 112ms/epoch - 7ms/step\n", "Epoch 153/400\n", "16/16 - 0s - loss: 0.3110 - accuracy: 0.9120 - val_loss: 0.3223 - val_accuracy: 0.9180 - 134ms/epoch - 8ms/step\n", "Epoch 154/400\n", "16/16 - 0s - loss: 0.3095 - accuracy: 0.9140 - val_loss: 0.3207 - val_accuracy: 0.9200 - 87ms/epoch - 5ms/step\n", "Epoch 155/400\n", "16/16 - 0s - loss: 0.3080 - accuracy: 0.9160 - val_loss: 0.3192 - val_accuracy: 0.9240 - 116ms/epoch - 7ms/step\n", "Epoch 156/400\n", "16/16 - 0s - loss: 0.3065 - accuracy: 0.9200 - val_loss: 0.3177 - val_accuracy: 0.9260 - 114ms/epoch - 7ms/step\n", "Epoch 157/400\n", "16/16 - 0s - loss: 0.3050 - accuracy: 0.9200 - val_loss: 0.3162 - val_accuracy: 0.9260 - 71ms/epoch - 4ms/step\n", "Epoch 158/400\n", "16/16 - 0s - loss: 0.3035 - accuracy: 0.9200 - val_loss: 0.3146 - val_accuracy: 0.9300 - 79ms/epoch - 5ms/step\n", "Epoch 159/400\n", "16/16 - 0s - loss: 0.3020 - accuracy: 0.9220 - val_loss: 0.3131 - val_accuracy: 0.9320 - 70ms/epoch - 4ms/step\n", "Epoch 160/400\n", "16/16 - 0s - loss: 0.3005 - accuracy: 0.9240 - val_loss: 0.3115 - val_accuracy: 0.9320 - 82ms/epoch - 5ms/step\n", "Epoch 161/400\n", "16/16 - 0s - loss: 0.2990 - accuracy: 0.9260 - val_loss: 0.3101 - val_accuracy: 0.9320 - 117ms/epoch - 7ms/step\n", "Epoch 162/400\n", "16/16 - 0s - loss: 0.2976 - accuracy: 0.9260 - val_loss: 0.3085 - val_accuracy: 0.9360 - 114ms/epoch - 7ms/step\n", "Epoch 163/400\n", "16/16 - 0s - loss: 0.2961 - accuracy: 0.9260 - val_loss: 0.3071 - val_accuracy: 0.9360 - 115ms/epoch - 7ms/step\n", "Epoch 164/400\n", "16/16 - 0s - loss: 0.2946 - accuracy: 0.9260 - val_loss: 0.3056 - val_accuracy: 0.9360 - 125ms/epoch - 8ms/step\n", "Epoch 165/400\n", "16/16 - 0s - loss: 0.2932 - accuracy: 0.9300 - val_loss: 0.3041 - val_accuracy: 0.9380 - 75ms/epoch - 5ms/step\n", "Epoch 166/400\n", "16/16 - 0s - loss: 0.2917 - accuracy: 0.9320 - val_loss: 0.3027 - val_accuracy: 0.9380 - 74ms/epoch - 5ms/step\n", "Epoch 167/400\n", "16/16 - 0s - loss: 0.2903 - accuracy: 0.9340 - val_loss: 0.3012 - val_accuracy: 0.9380 - 113ms/epoch - 7ms/step\n", "Epoch 168/400\n", "16/16 - 0s - loss: 0.2888 - accuracy: 0.9360 - val_loss: 0.2997 - val_accuracy: 0.9380 - 115ms/epoch - 7ms/step\n", "Epoch 169/400\n", "16/16 - 0s - loss: 0.2873 - accuracy: 0.9360 - val_loss: 0.2982 - val_accuracy: 0.9380 - 114ms/epoch - 7ms/step\n", "Epoch 170/400\n", "16/16 - 0s - loss: 0.2859 - accuracy: 0.9380 - val_loss: 0.2968 - val_accuracy: 0.9380 - 111ms/epoch - 7ms/step\n", "Epoch 171/400\n", "16/16 - 0s - loss: 0.2845 - accuracy: 0.9420 - val_loss: 0.2952 - val_accuracy: 0.9420 - 112ms/epoch - 7ms/step\n", "Epoch 172/400\n", "16/16 - 0s - loss: 0.2830 - accuracy: 0.9460 - val_loss: 0.2938 - val_accuracy: 0.9420 - 81ms/epoch - 5ms/step\n", "Epoch 173/400\n", "16/16 - 0s - loss: 0.2815 - accuracy: 0.9480 - val_loss: 0.2923 - val_accuracy: 0.9440 - 117ms/epoch - 7ms/step\n", "Epoch 174/400\n", "16/16 - 0s - loss: 0.2801 - accuracy: 0.9480 - val_loss: 0.2907 - val_accuracy: 0.9440 - 77ms/epoch - 5ms/step\n", "Epoch 175/400\n", "16/16 - 0s - loss: 0.2786 - accuracy: 0.9500 - val_loss: 0.2893 - val_accuracy: 0.9440 - 116ms/epoch - 7ms/step\n", "Epoch 176/400\n", "16/16 - 0s - loss: 0.2772 - accuracy: 0.9500 - val_loss: 0.2878 - val_accuracy: 0.9440 - 113ms/epoch - 7ms/step\n", "Epoch 177/400\n", "16/16 - 0s - loss: 0.2757 - accuracy: 0.9540 - val_loss: 0.2863 - val_accuracy: 0.9460 - 118ms/epoch - 7ms/step\n", "Epoch 178/400\n", "16/16 - 0s - loss: 0.2743 - accuracy: 0.9560 - val_loss: 0.2848 - val_accuracy: 0.9480 - 76ms/epoch - 5ms/step\n", "Epoch 179/400\n", "16/16 - 0s - loss: 0.2729 - accuracy: 0.9580 - val_loss: 0.2833 - val_accuracy: 0.9480 - 114ms/epoch - 7ms/step\n", "Epoch 180/400\n", "16/16 - 0s - loss: 0.2714 - accuracy: 0.9580 - val_loss: 0.2820 - val_accuracy: 0.9480 - 113ms/epoch - 7ms/step\n", "Epoch 181/400\n", "16/16 - 0s - loss: 0.2700 - accuracy: 0.9580 - val_loss: 0.2804 - val_accuracy: 0.9480 - 76ms/epoch - 5ms/step\n", "Epoch 182/400\n", "16/16 - 0s - loss: 0.2685 - accuracy: 0.9620 - val_loss: 0.2789 - val_accuracy: 0.9540 - 78ms/epoch - 5ms/step\n", "Epoch 183/400\n", "16/16 - 0s - loss: 0.2671 - accuracy: 0.9620 - val_loss: 0.2774 - val_accuracy: 0.9540 - 83ms/epoch - 5ms/step\n", "Epoch 184/400\n", "16/16 - 0s - loss: 0.2657 - accuracy: 0.9660 - val_loss: 0.2759 - val_accuracy: 0.9540 - 72ms/epoch - 4ms/step\n", "Epoch 185/400\n", "16/16 - 0s - loss: 0.2642 - accuracy: 0.9660 - val_loss: 0.2744 - val_accuracy: 0.9540 - 77ms/epoch - 5ms/step\n", "Epoch 186/400\n", "16/16 - 0s - loss: 0.2628 - accuracy: 0.9660 - val_loss: 0.2729 - val_accuracy: 0.9560 - 71ms/epoch - 4ms/step\n", "Epoch 187/400\n", "16/16 - 0s - loss: 0.2613 - accuracy: 0.9660 - val_loss: 0.2714 - val_accuracy: 0.9580 - 73ms/epoch - 5ms/step\n", "Epoch 188/400\n", "16/16 - 0s - loss: 0.2599 - accuracy: 0.9680 - val_loss: 0.2699 - val_accuracy: 0.9580 - 77ms/epoch - 5ms/step\n", "Epoch 189/400\n", "16/16 - 0s - loss: 0.2585 - accuracy: 0.9680 - val_loss: 0.2684 - val_accuracy: 0.9580 - 75ms/epoch - 5ms/step\n", "Epoch 190/400\n", "16/16 - 0s - loss: 0.2570 - accuracy: 0.9680 - val_loss: 0.2669 - val_accuracy: 0.9580 - 117ms/epoch - 7ms/step\n", "Epoch 191/400\n", "16/16 - 0s - loss: 0.2556 - accuracy: 0.9680 - val_loss: 0.2654 - val_accuracy: 0.9620 - 111ms/epoch - 7ms/step\n", "Epoch 192/400\n", "16/16 - 0s - loss: 0.2542 - accuracy: 0.9680 - val_loss: 0.2638 - val_accuracy: 0.9640 - 79ms/epoch - 5ms/step\n", "Epoch 193/400\n", "16/16 - 0s - loss: 0.2527 - accuracy: 0.9680 - val_loss: 0.2624 - val_accuracy: 0.9640 - 80ms/epoch - 5ms/step\n", "Epoch 194/400\n", "16/16 - 0s - loss: 0.2512 - accuracy: 0.9680 - val_loss: 0.2609 - val_accuracy: 0.9640 - 119ms/epoch - 7ms/step\n", "Epoch 195/400\n", "16/16 - 0s - loss: 0.2498 - accuracy: 0.9700 - val_loss: 0.2593 - val_accuracy: 0.9640 - 122ms/epoch - 8ms/step\n", "Epoch 196/400\n", "16/16 - 0s - loss: 0.2483 - accuracy: 0.9720 - val_loss: 0.2578 - val_accuracy: 0.9640 - 80ms/epoch - 5ms/step\n", "Epoch 197/400\n", "16/16 - 0s - loss: 0.2468 - accuracy: 0.9720 - val_loss: 0.2562 - val_accuracy: 0.9660 - 74ms/epoch - 5ms/step\n", "Epoch 198/400\n", "16/16 - 0s - loss: 0.2454 - accuracy: 0.9760 - val_loss: 0.2548 - val_accuracy: 0.9660 - 113ms/epoch - 7ms/step\n", "Epoch 199/400\n", "16/16 - 0s - loss: 0.2440 - accuracy: 0.9760 - val_loss: 0.2530 - val_accuracy: 0.9680 - 74ms/epoch - 5ms/step\n", "Epoch 200/400\n", "16/16 - 0s - loss: 0.2424 - accuracy: 0.9780 - val_loss: 0.2516 - val_accuracy: 0.9680 - 70ms/epoch - 4ms/step\n", "Epoch 201/400\n", "16/16 - 0s - loss: 0.2410 - accuracy: 0.9780 - val_loss: 0.2500 - val_accuracy: 0.9680 - 71ms/epoch - 4ms/step\n", "Epoch 202/400\n", "16/16 - 0s - loss: 0.2395 - accuracy: 0.9780 - val_loss: 0.2485 - val_accuracy: 0.9700 - 113ms/epoch - 7ms/step\n", "Epoch 203/400\n", "16/16 - 0s - loss: 0.2380 - accuracy: 0.9780 - val_loss: 0.2470 - val_accuracy: 0.9720 - 112ms/epoch - 7ms/step\n", "Epoch 204/400\n", "16/16 - 0s - loss: 0.2366 - accuracy: 0.9800 - val_loss: 0.2455 - val_accuracy: 0.9740 - 79ms/epoch - 5ms/step\n", "Epoch 205/400\n", "16/16 - 0s - loss: 0.2351 - accuracy: 0.9820 - val_loss: 0.2440 - val_accuracy: 0.9740 - 84ms/epoch - 5ms/step\n", "Epoch 206/400\n", "16/16 - 0s - loss: 0.2336 - accuracy: 0.9820 - val_loss: 0.2425 - val_accuracy: 0.9740 - 114ms/epoch - 7ms/step\n", "Epoch 207/400\n", "16/16 - 0s - loss: 0.2322 - accuracy: 0.9820 - val_loss: 0.2409 - val_accuracy: 0.9740 - 115ms/epoch - 7ms/step\n", "Epoch 208/400\n", "16/16 - 0s - loss: 0.2307 - accuracy: 0.9820 - val_loss: 0.2394 - val_accuracy: 0.9760 - 73ms/epoch - 5ms/step\n", "Epoch 209/400\n", "16/16 - 0s - loss: 0.2292 - accuracy: 0.9840 - val_loss: 0.2378 - val_accuracy: 0.9760 - 85ms/epoch - 5ms/step\n", "Epoch 210/400\n", "16/16 - 0s - loss: 0.2277 - accuracy: 0.9840 - val_loss: 0.2363 - val_accuracy: 0.9760 - 110ms/epoch - 7ms/step\n", "Epoch 211/400\n", "16/16 - 0s - loss: 0.2262 - accuracy: 0.9840 - val_loss: 0.2346 - val_accuracy: 0.9760 - 73ms/epoch - 5ms/step\n", "Epoch 212/400\n", "16/16 - 0s - loss: 0.2246 - accuracy: 0.9880 - val_loss: 0.2331 - val_accuracy: 0.9800 - 112ms/epoch - 7ms/step\n", "Epoch 213/400\n", "16/16 - 0s - loss: 0.2231 - accuracy: 0.9880 - val_loss: 0.2315 - val_accuracy: 0.9820 - 75ms/epoch - 5ms/step\n", "Epoch 214/400\n", "16/16 - 0s - loss: 0.2216 - accuracy: 0.9880 - val_loss: 0.2300 - val_accuracy: 0.9820 - 113ms/epoch - 7ms/step\n", "Epoch 215/400\n", "16/16 - 0s - loss: 0.2201 - accuracy: 0.9900 - val_loss: 0.2283 - val_accuracy: 0.9820 - 95ms/epoch - 6ms/step\n", "Epoch 216/400\n", "16/16 - 0s - loss: 0.2185 - accuracy: 0.9900 - val_loss: 0.2268 - val_accuracy: 0.9840 - 82ms/epoch - 5ms/step\n", "Epoch 217/400\n", "16/16 - 0s - loss: 0.2170 - accuracy: 0.9900 - val_loss: 0.2252 - val_accuracy: 0.9840 - 79ms/epoch - 5ms/step\n", "Epoch 218/400\n", "16/16 - 0s - loss: 0.2155 - accuracy: 0.9920 - val_loss: 0.2237 - val_accuracy: 0.9860 - 73ms/epoch - 5ms/step\n", "Epoch 219/400\n", "16/16 - 0s - loss: 0.2140 - accuracy: 0.9920 - val_loss: 0.2221 - val_accuracy: 0.9880 - 74ms/epoch - 5ms/step\n", "Epoch 220/400\n", "16/16 - 0s - loss: 0.2125 - accuracy: 0.9920 - val_loss: 0.2205 - val_accuracy: 0.9880 - 115ms/epoch - 7ms/step\n", "Epoch 221/400\n", "16/16 - 0s - loss: 0.2109 - accuracy: 0.9920 - val_loss: 0.2190 - val_accuracy: 0.9880 - 118ms/epoch - 7ms/step\n", "Epoch 222/400\n", "16/16 - 0s - loss: 0.2094 - accuracy: 0.9920 - val_loss: 0.2175 - val_accuracy: 0.9880 - 124ms/epoch - 8ms/step\n", "Epoch 223/400\n", "16/16 - 0s - loss: 0.2079 - accuracy: 0.9920 - val_loss: 0.2159 - val_accuracy: 0.9880 - 75ms/epoch - 5ms/step\n", "Epoch 224/400\n", "16/16 - 0s - loss: 0.2064 - accuracy: 0.9920 - val_loss: 0.2144 - val_accuracy: 0.9880 - 77ms/epoch - 5ms/step\n", "Epoch 225/400\n", "16/16 - 0s - loss: 0.2050 - accuracy: 0.9920 - val_loss: 0.2130 - val_accuracy: 0.9880 - 73ms/epoch - 5ms/step\n", "Epoch 226/400\n", "16/16 - 0s - loss: 0.2034 - accuracy: 0.9940 - val_loss: 0.2114 - val_accuracy: 0.9880 - 85ms/epoch - 5ms/step\n", "Epoch 227/400\n", "16/16 - 0s - loss: 0.2020 - accuracy: 0.9940 - val_loss: 0.2098 - val_accuracy: 0.9880 - 119ms/epoch - 7ms/step\n", "Epoch 228/400\n", "16/16 - 0s - loss: 0.2005 - accuracy: 0.9940 - val_loss: 0.2082 - val_accuracy: 0.9900 - 80ms/epoch - 5ms/step\n", "Epoch 229/400\n", "16/16 - 0s - loss: 0.1990 - accuracy: 0.9940 - val_loss: 0.2068 - val_accuracy: 0.9900 - 79ms/epoch - 5ms/step\n", "Epoch 230/400\n", "16/16 - 0s - loss: 0.1975 - accuracy: 0.9940 - val_loss: 0.2052 - val_accuracy: 0.9900 - 114ms/epoch - 7ms/step\n", "Epoch 231/400\n", "16/16 - 0s - loss: 0.1960 - accuracy: 0.9940 - val_loss: 0.2038 - val_accuracy: 0.9920 - 112ms/epoch - 7ms/step\n", "Epoch 232/400\n", "16/16 - 0s - loss: 0.1945 - accuracy: 0.9940 - val_loss: 0.2022 - val_accuracy: 0.9920 - 77ms/epoch - 5ms/step\n", "Epoch 233/400\n", "16/16 - 0s - loss: 0.1930 - accuracy: 0.9940 - val_loss: 0.2007 - val_accuracy: 0.9920 - 116ms/epoch - 7ms/step\n", "Epoch 234/400\n", "16/16 - 0s - loss: 0.1915 - accuracy: 0.9960 - val_loss: 0.1992 - val_accuracy: 0.9920 - 73ms/epoch - 5ms/step\n", "Epoch 235/400\n", "16/16 - 0s - loss: 0.1900 - accuracy: 0.9960 - val_loss: 0.1976 - val_accuracy: 0.9920 - 115ms/epoch - 7ms/step\n", "Epoch 236/400\n", "16/16 - 0s - loss: 0.1885 - accuracy: 0.9960 - val_loss: 0.1961 - val_accuracy: 0.9920 - 90ms/epoch - 6ms/step\n", "Epoch 237/400\n", "16/16 - 0s - loss: 0.1871 - accuracy: 0.9960 - val_loss: 0.1946 - val_accuracy: 0.9920 - 73ms/epoch - 5ms/step\n", "Epoch 238/400\n", "16/16 - 0s - loss: 0.1856 - accuracy: 0.9960 - val_loss: 0.1931 - val_accuracy: 0.9920 - 116ms/epoch - 7ms/step\n", "Epoch 239/400\n", "16/16 - 0s - loss: 0.1842 - accuracy: 0.9960 - val_loss: 0.1917 - val_accuracy: 0.9920 - 76ms/epoch - 5ms/step\n", "Epoch 240/400\n", "16/16 - 0s - loss: 0.1828 - accuracy: 0.9960 - val_loss: 0.1902 - val_accuracy: 0.9920 - 113ms/epoch - 7ms/step\n", "Epoch 241/400\n", "16/16 - 0s - loss: 0.1814 - accuracy: 0.9960 - val_loss: 0.1889 - val_accuracy: 0.9920 - 72ms/epoch - 4ms/step\n", "Epoch 242/400\n", "16/16 - 0s - loss: 0.1800 - accuracy: 0.9960 - val_loss: 0.1874 - val_accuracy: 0.9920 - 74ms/epoch - 5ms/step\n", "Epoch 243/400\n", "16/16 - 0s - loss: 0.1786 - accuracy: 0.9960 - val_loss: 0.1860 - val_accuracy: 0.9920 - 76ms/epoch - 5ms/step\n", "Epoch 244/400\n", "16/16 - 0s - loss: 0.1772 - accuracy: 0.9960 - val_loss: 0.1846 - val_accuracy: 0.9920 - 117ms/epoch - 7ms/step\n", "Epoch 245/400\n", "16/16 - 0s - loss: 0.1758 - accuracy: 0.9960 - val_loss: 0.1831 - val_accuracy: 0.9920 - 111ms/epoch - 7ms/step\n", "Epoch 246/400\n", "16/16 - 0s - loss: 0.1744 - accuracy: 0.9960 - val_loss: 0.1816 - val_accuracy: 0.9920 - 118ms/epoch - 7ms/step\n", "Epoch 247/400\n", "16/16 - 0s - loss: 0.1730 - accuracy: 0.9960 - val_loss: 0.1802 - val_accuracy: 0.9920 - 116ms/epoch - 7ms/step\n", "Epoch 248/400\n", "16/16 - 0s - loss: 0.1716 - accuracy: 0.9960 - val_loss: 0.1789 - val_accuracy: 0.9920 - 119ms/epoch - 7ms/step\n", "Epoch 249/400\n", "16/16 - 0s - loss: 0.1703 - accuracy: 0.9960 - val_loss: 0.1774 - val_accuracy: 0.9920 - 115ms/epoch - 7ms/step\n", "Epoch 250/400\n", "16/16 - 0s - loss: 0.1688 - accuracy: 0.9960 - val_loss: 0.1761 - val_accuracy: 0.9920 - 70ms/epoch - 4ms/step\n", "Epoch 251/400\n", "16/16 - 0s - loss: 0.1674 - accuracy: 0.9960 - val_loss: 0.1746 - val_accuracy: 0.9920 - 113ms/epoch - 7ms/step\n", "Epoch 252/400\n", "16/16 - 0s - loss: 0.1661 - accuracy: 0.9960 - val_loss: 0.1732 - val_accuracy: 0.9920 - 116ms/epoch - 7ms/step\n", "Epoch 253/400\n", "16/16 - 0s - loss: 0.1647 - accuracy: 0.9960 - val_loss: 0.1718 - val_accuracy: 0.9920 - 115ms/epoch - 7ms/step\n", "Epoch 254/400\n", "16/16 - 0s - loss: 0.1633 - accuracy: 0.9960 - val_loss: 0.1705 - val_accuracy: 0.9920 - 115ms/epoch - 7ms/step\n", "Epoch 255/400\n", "16/16 - 0s - loss: 0.1619 - accuracy: 0.9960 - val_loss: 0.1691 - val_accuracy: 0.9920 - 88ms/epoch - 5ms/step\n", "Epoch 256/400\n", "16/16 - 0s - loss: 0.1605 - accuracy: 0.9960 - val_loss: 0.1676 - val_accuracy: 0.9920 - 122ms/epoch - 8ms/step\n", "Epoch 257/400\n", "16/16 - 0s - loss: 0.1591 - accuracy: 0.9960 - val_loss: 0.1663 - val_accuracy: 0.9940 - 113ms/epoch - 7ms/step\n", "Epoch 258/400\n", "16/16 - 0s - loss: 0.1578 - accuracy: 0.9960 - val_loss: 0.1649 - val_accuracy: 0.9940 - 75ms/epoch - 5ms/step\n", "Epoch 259/400\n", "16/16 - 0s - loss: 0.1564 - accuracy: 0.9960 - val_loss: 0.1635 - val_accuracy: 0.9940 - 74ms/epoch - 5ms/step\n", "Epoch 260/400\n", "16/16 - 0s - loss: 0.1550 - accuracy: 0.9960 - val_loss: 0.1622 - val_accuracy: 0.9940 - 114ms/epoch - 7ms/step\n", "Epoch 261/400\n", "16/16 - 0s - loss: 0.1537 - accuracy: 0.9960 - val_loss: 0.1608 - val_accuracy: 0.9940 - 118ms/epoch - 7ms/step\n", "Epoch 262/400\n", "16/16 - 0s - loss: 0.1524 - accuracy: 0.9960 - val_loss: 0.1594 - val_accuracy: 0.9940 - 77ms/epoch - 5ms/step\n", "Epoch 263/400\n", "16/16 - 0s - loss: 0.1510 - accuracy: 0.9960 - val_loss: 0.1581 - val_accuracy: 0.9940 - 114ms/epoch - 7ms/step\n", "Epoch 264/400\n", "16/16 - 0s - loss: 0.1497 - accuracy: 0.9960 - val_loss: 0.1567 - val_accuracy: 0.9940 - 116ms/epoch - 7ms/step\n", "Epoch 265/400\n", "16/16 - 0s - loss: 0.1484 - accuracy: 0.9960 - val_loss: 0.1554 - val_accuracy: 0.9940 - 86ms/epoch - 5ms/step\n", "Epoch 266/400\n", "16/16 - 0s - loss: 0.1470 - accuracy: 0.9960 - val_loss: 0.1539 - val_accuracy: 0.9940 - 77ms/epoch - 5ms/step\n", "Epoch 267/400\n", "16/16 - 0s - loss: 0.1457 - accuracy: 0.9980 - val_loss: 0.1525 - val_accuracy: 0.9940 - 121ms/epoch - 8ms/step\n", "Epoch 268/400\n", "16/16 - 0s - loss: 0.1444 - accuracy: 1.0000 - val_loss: 0.1512 - val_accuracy: 0.9940 - 119ms/epoch - 7ms/step\n", "Epoch 269/400\n", "16/16 - 0s - loss: 0.1430 - accuracy: 1.0000 - val_loss: 0.1499 - val_accuracy: 0.9940 - 113ms/epoch - 7ms/step\n", "Epoch 270/400\n", "16/16 - 0s - loss: 0.1417 - accuracy: 1.0000 - val_loss: 0.1486 - val_accuracy: 0.9940 - 74ms/epoch - 5ms/step\n", "Epoch 271/400\n", "16/16 - 0s - loss: 0.1403 - accuracy: 1.0000 - val_loss: 0.1472 - val_accuracy: 0.9940 - 79ms/epoch - 5ms/step\n", "Epoch 272/400\n", "16/16 - 0s - loss: 0.1390 - accuracy: 1.0000 - val_loss: 0.1459 - val_accuracy: 0.9940 - 86ms/epoch - 5ms/step\n", "Epoch 273/400\n", "16/16 - 0s - loss: 0.1377 - accuracy: 1.0000 - val_loss: 0.1445 - val_accuracy: 0.9940 - 117ms/epoch - 7ms/step\n", "Epoch 274/400\n", "16/16 - 0s - loss: 0.1363 - accuracy: 1.0000 - val_loss: 0.1431 - val_accuracy: 0.9940 - 95ms/epoch - 6ms/step\n", "Epoch 275/400\n", "16/16 - 0s - loss: 0.1350 - accuracy: 1.0000 - val_loss: 0.1418 - val_accuracy: 0.9940 - 94ms/epoch - 6ms/step\n", "Epoch 276/400\n", "16/16 - 0s - loss: 0.1337 - accuracy: 1.0000 - val_loss: 0.1405 - val_accuracy: 0.9960 - 121ms/epoch - 8ms/step\n", "Epoch 277/400\n", "16/16 - 0s - loss: 0.1324 - accuracy: 1.0000 - val_loss: 0.1391 - val_accuracy: 0.9960 - 119ms/epoch - 7ms/step\n", "Epoch 278/400\n", "16/16 - 0s - loss: 0.1311 - accuracy: 1.0000 - val_loss: 0.1378 - val_accuracy: 0.9960 - 75ms/epoch - 5ms/step\n", "Epoch 279/400\n", "16/16 - 0s - loss: 0.1298 - accuracy: 1.0000 - val_loss: 0.1364 - val_accuracy: 0.9960 - 70ms/epoch - 4ms/step\n", "Epoch 280/400\n", "16/16 - 0s - loss: 0.1285 - accuracy: 1.0000 - val_loss: 0.1351 - val_accuracy: 0.9960 - 121ms/epoch - 8ms/step\n", "Epoch 281/400\n", "16/16 - 0s - loss: 0.1271 - accuracy: 1.0000 - val_loss: 0.1338 - val_accuracy: 0.9960 - 73ms/epoch - 5ms/step\n", "Epoch 282/400\n", "16/16 - 0s - loss: 0.1259 - accuracy: 1.0000 - val_loss: 0.1325 - val_accuracy: 0.9960 - 80ms/epoch - 5ms/step\n", "Epoch 283/400\n", "16/16 - 0s - loss: 0.1246 - accuracy: 1.0000 - val_loss: 0.1312 - val_accuracy: 0.9960 - 81ms/epoch - 5ms/step\n", "Epoch 284/400\n", "16/16 - 0s - loss: 0.1233 - accuracy: 1.0000 - val_loss: 0.1299 - val_accuracy: 0.9960 - 75ms/epoch - 5ms/step\n", "Epoch 285/400\n", "16/16 - 0s - loss: 0.1221 - accuracy: 1.0000 - val_loss: 0.1286 - val_accuracy: 0.9960 - 117ms/epoch - 7ms/step\n", "Epoch 286/400\n", "16/16 - 0s - loss: 0.1208 - accuracy: 1.0000 - val_loss: 0.1273 - val_accuracy: 0.9960 - 116ms/epoch - 7ms/step\n", "Epoch 287/400\n", "16/16 - 0s - loss: 0.1195 - accuracy: 1.0000 - val_loss: 0.1260 - val_accuracy: 0.9960 - 111ms/epoch - 7ms/step\n", "Epoch 288/400\n", "16/16 - 0s - loss: 0.1183 - accuracy: 1.0000 - val_loss: 0.1247 - val_accuracy: 0.9980 - 120ms/epoch - 7ms/step\n", "Epoch 289/400\n", "16/16 - 0s - loss: 0.1170 - accuracy: 1.0000 - val_loss: 0.1235 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 290/400\n", "16/16 - 0s - loss: 0.1158 - accuracy: 1.0000 - val_loss: 0.1223 - val_accuracy: 0.9980 - 77ms/epoch - 5ms/step\n", "Epoch 291/400\n", "16/16 - 0s - loss: 0.1145 - accuracy: 1.0000 - val_loss: 0.1211 - val_accuracy: 0.9980 - 74ms/epoch - 5ms/step\n", "Epoch 292/400\n", "16/16 - 0s - loss: 0.1133 - accuracy: 1.0000 - val_loss: 0.1198 - val_accuracy: 0.9980 - 78ms/epoch - 5ms/step\n", "Epoch 293/400\n", "16/16 - 0s - loss: 0.1120 - accuracy: 1.0000 - val_loss: 0.1186 - val_accuracy: 0.9980 - 76ms/epoch - 5ms/step\n", "Epoch 294/400\n", "16/16 - 0s - loss: 0.1108 - accuracy: 1.0000 - val_loss: 0.1173 - val_accuracy: 0.9980 - 82ms/epoch - 5ms/step\n", "Epoch 295/400\n", "16/16 - 0s - loss: 0.1096 - accuracy: 1.0000 - val_loss: 0.1162 - val_accuracy: 0.9980 - 84ms/epoch - 5ms/step\n", "Epoch 296/400\n", "16/16 - 0s - loss: 0.1084 - accuracy: 1.0000 - val_loss: 0.1149 - val_accuracy: 0.9980 - 124ms/epoch - 8ms/step\n", "Epoch 297/400\n", "16/16 - 0s - loss: 0.1071 - accuracy: 1.0000 - val_loss: 0.1137 - val_accuracy: 0.9980 - 120ms/epoch - 7ms/step\n", "Epoch 298/400\n", "16/16 - 0s - loss: 0.1059 - accuracy: 1.0000 - val_loss: 0.1125 - val_accuracy: 0.9980 - 75ms/epoch - 5ms/step\n", "Epoch 299/400\n", "16/16 - 0s - loss: 0.1047 - accuracy: 1.0000 - val_loss: 0.1113 - val_accuracy: 0.9980 - 114ms/epoch - 7ms/step\n", "Epoch 300/400\n", "16/16 - 0s - loss: 0.1035 - accuracy: 1.0000 - val_loss: 0.1101 - val_accuracy: 0.9980 - 76ms/epoch - 5ms/step\n", "Epoch 301/400\n", "16/16 - 0s - loss: 0.1023 - accuracy: 1.0000 - val_loss: 0.1090 - val_accuracy: 0.9980 - 79ms/epoch - 5ms/step\n", "Epoch 302/400\n", "16/16 - 0s - loss: 0.1012 - accuracy: 1.0000 - val_loss: 0.1078 - val_accuracy: 0.9980 - 117ms/epoch - 7ms/step\n", "Epoch 303/400\n", "16/16 - 0s - loss: 0.1000 - accuracy: 1.0000 - val_loss: 0.1067 - val_accuracy: 0.9980 - 72ms/epoch - 5ms/step\n", "Epoch 304/400\n", "16/16 - 0s - loss: 0.0989 - accuracy: 1.0000 - val_loss: 0.1056 - val_accuracy: 1.0000 - 116ms/epoch - 7ms/step\n", "Epoch 305/400\n", "16/16 - 0s - loss: 0.0977 - accuracy: 1.0000 - val_loss: 0.1045 - val_accuracy: 1.0000 - 115ms/epoch - 7ms/step\n", "Epoch 306/400\n", "16/16 - 0s - loss: 0.0966 - accuracy: 1.0000 - val_loss: 0.1034 - val_accuracy: 1.0000 - 88ms/epoch - 6ms/step\n", "Epoch 307/400\n", "16/16 - 0s - loss: 0.0955 - accuracy: 1.0000 - val_loss: 0.1023 - val_accuracy: 1.0000 - 120ms/epoch - 8ms/step\n", "Epoch 308/400\n", "16/16 - 0s - loss: 0.0944 - accuracy: 1.0000 - val_loss: 0.1012 - val_accuracy: 1.0000 - 113ms/epoch - 7ms/step\n", "Epoch 309/400\n", "16/16 - 0s - loss: 0.0933 - accuracy: 1.0000 - val_loss: 0.1000 - val_accuracy: 1.0000 - 74ms/epoch - 5ms/step\n", "Epoch 310/400\n", "16/16 - 0s - loss: 0.0922 - accuracy: 1.0000 - val_loss: 0.0990 - val_accuracy: 1.0000 - 81ms/epoch - 5ms/step\n", "Epoch 311/400\n", "16/16 - 0s - loss: 0.0911 - accuracy: 1.0000 - val_loss: 0.0980 - val_accuracy: 1.0000 - 111ms/epoch - 7ms/step\n", "Epoch 312/400\n", "16/16 - 0s - loss: 0.0901 - accuracy: 1.0000 - val_loss: 0.0969 - val_accuracy: 1.0000 - 78ms/epoch - 5ms/step\n", "Epoch 313/400\n", "16/16 - 0s - loss: 0.0890 - accuracy: 1.0000 - val_loss: 0.0959 - val_accuracy: 1.0000 - 117ms/epoch - 7ms/step\n", "Epoch 314/400\n", "16/16 - 0s - loss: 0.0879 - accuracy: 1.0000 - val_loss: 0.0948 - val_accuracy: 1.0000 - 121ms/epoch - 8ms/step\n", "Epoch 315/400\n", "16/16 - 0s - loss: 0.0869 - accuracy: 1.0000 - val_loss: 0.0938 - val_accuracy: 1.0000 - 74ms/epoch - 5ms/step\n", "Epoch 316/400\n", "16/16 - 0s - loss: 0.0859 - accuracy: 1.0000 - val_loss: 0.0928 - val_accuracy: 1.0000 - 87ms/epoch - 5ms/step\n", "Epoch 317/400\n", "16/16 - 0s - loss: 0.0849 - accuracy: 1.0000 - val_loss: 0.0918 - val_accuracy: 1.0000 - 74ms/epoch - 5ms/step\n", "Epoch 318/400\n", "16/16 - 0s - loss: 0.0840 - accuracy: 1.0000 - val_loss: 0.0909 - val_accuracy: 1.0000 - 76ms/epoch - 5ms/step\n", "Epoch 319/400\n", "16/16 - 0s - loss: 0.0830 - accuracy: 1.0000 - val_loss: 0.0899 - val_accuracy: 1.0000 - 77ms/epoch - 5ms/step\n", "Epoch 320/400\n", "16/16 - 0s - loss: 0.0820 - accuracy: 1.0000 - val_loss: 0.0890 - val_accuracy: 1.0000 - 117ms/epoch - 7ms/step\n", "Epoch 321/400\n", "16/16 - 0s - loss: 0.0811 - accuracy: 1.0000 - val_loss: 0.0880 - val_accuracy: 1.0000 - 113ms/epoch - 7ms/step\n", "Epoch 322/400\n", "16/16 - 0s - loss: 0.0802 - accuracy: 1.0000 - val_loss: 0.0872 - val_accuracy: 1.0000 - 118ms/epoch - 7ms/step\n", "Epoch 323/400\n", "16/16 - 0s - loss: 0.0792 - accuracy: 1.0000 - val_loss: 0.0863 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 324/400\n", "16/16 - 0s - loss: 0.0783 - accuracy: 1.0000 - val_loss: 0.0854 - val_accuracy: 1.0000 - 116ms/epoch - 7ms/step\n", "Epoch 325/400\n", "16/16 - 0s - loss: 0.0774 - accuracy: 1.0000 - val_loss: 0.0845 - val_accuracy: 1.0000 - 81ms/epoch - 5ms/step\n", "Epoch 326/400\n", "16/16 - 0s - loss: 0.0765 - accuracy: 1.0000 - val_loss: 0.0836 - val_accuracy: 1.0000 - 117ms/epoch - 7ms/step\n", "Epoch 327/400\n", "16/16 - 0s - loss: 0.0756 - accuracy: 1.0000 - val_loss: 0.0828 - val_accuracy: 1.0000 - 77ms/epoch - 5ms/step\n", "Epoch 328/400\n", "16/16 - 0s - loss: 0.0748 - accuracy: 1.0000 - val_loss: 0.0819 - val_accuracy: 1.0000 - 81ms/epoch - 5ms/step\n", "Epoch 329/400\n", "16/16 - 0s - loss: 0.0739 - accuracy: 1.0000 - val_loss: 0.0811 - val_accuracy: 1.0000 - 76ms/epoch - 5ms/step\n", "Epoch 330/400\n", "16/16 - 0s - loss: 0.0730 - accuracy: 1.0000 - val_loss: 0.0803 - val_accuracy: 1.0000 - 114ms/epoch - 7ms/step\n", "Epoch 331/400\n", "16/16 - 0s - loss: 0.0722 - accuracy: 1.0000 - val_loss: 0.0795 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 332/400\n", "16/16 - 0s - loss: 0.0714 - accuracy: 1.0000 - val_loss: 0.0787 - val_accuracy: 1.0000 - 122ms/epoch - 8ms/step\n", "Epoch 333/400\n", "16/16 - 0s - loss: 0.0706 - accuracy: 1.0000 - val_loss: 0.0779 - val_accuracy: 1.0000 - 80ms/epoch - 5ms/step\n", "Epoch 334/400\n", "16/16 - 0s - loss: 0.0698 - accuracy: 1.0000 - val_loss: 0.0771 - val_accuracy: 1.0000 - 77ms/epoch - 5ms/step\n", "Epoch 335/400\n", "16/16 - 0s - loss: 0.0690 - accuracy: 1.0000 - val_loss: 0.0763 - val_accuracy: 1.0000 - 118ms/epoch - 7ms/step\n", "Epoch 336/400\n", "16/16 - 0s - loss: 0.0682 - accuracy: 1.0000 - val_loss: 0.0756 - val_accuracy: 1.0000 - 86ms/epoch - 5ms/step\n", "Epoch 337/400\n", "16/16 - 0s - loss: 0.0675 - accuracy: 1.0000 - val_loss: 0.0749 - val_accuracy: 1.0000 - 82ms/epoch - 5ms/step\n", "Epoch 338/400\n", "16/16 - 0s - loss: 0.0667 - accuracy: 1.0000 - val_loss: 0.0742 - val_accuracy: 1.0000 - 119ms/epoch - 7ms/step\n", "Epoch 339/400\n", "16/16 - 0s - loss: 0.0660 - accuracy: 1.0000 - val_loss: 0.0734 - val_accuracy: 1.0000 - 78ms/epoch - 5ms/step\n", "Epoch 340/400\n", "16/16 - 0s - loss: 0.0653 - accuracy: 1.0000 - val_loss: 0.0727 - val_accuracy: 1.0000 - 123ms/epoch - 8ms/step\n", "Epoch 341/400\n", "16/16 - 0s - loss: 0.0646 - accuracy: 1.0000 - val_loss: 0.0720 - val_accuracy: 1.0000 - 76ms/epoch - 5ms/step\n", "Epoch 342/400\n", "16/16 - 0s - loss: 0.0639 - accuracy: 1.0000 - val_loss: 0.0713 - val_accuracy: 1.0000 - 84ms/epoch - 5ms/step\n", "Epoch 343/400\n", "16/16 - 0s - loss: 0.0631 - accuracy: 1.0000 - val_loss: 0.0706 - val_accuracy: 1.0000 - 117ms/epoch - 7ms/step\n", "Epoch 344/400\n", "16/16 - 0s - loss: 0.0625 - accuracy: 1.0000 - val_loss: 0.0700 - val_accuracy: 1.0000 - 88ms/epoch - 6ms/step\n", "Epoch 345/400\n", "16/16 - 0s - loss: 0.0618 - accuracy: 1.0000 - val_loss: 0.0693 - val_accuracy: 1.0000 - 119ms/epoch - 7ms/step\n", "Epoch 346/400\n", "16/16 - 0s - loss: 0.0611 - accuracy: 1.0000 - val_loss: 0.0686 - val_accuracy: 1.0000 - 113ms/epoch - 7ms/step\n", "Epoch 347/400\n", "16/16 - 0s - loss: 0.0605 - accuracy: 1.0000 - val_loss: 0.0680 - val_accuracy: 1.0000 - 76ms/epoch - 5ms/step\n", "Epoch 348/400\n", "16/16 - 0s - loss: 0.0598 - accuracy: 1.0000 - val_loss: 0.0673 - val_accuracy: 1.0000 - 109ms/epoch - 7ms/step\n", "Epoch 349/400\n", "16/16 - 0s - loss: 0.0592 - accuracy: 1.0000 - val_loss: 0.0667 - val_accuracy: 1.0000 - 78ms/epoch - 5ms/step\n", "Epoch 350/400\n", "16/16 - 0s - loss: 0.0585 - accuracy: 1.0000 - val_loss: 0.0661 - val_accuracy: 1.0000 - 70ms/epoch - 4ms/step\n", "Epoch 351/400\n", "16/16 - 0s - loss: 0.0579 - accuracy: 1.0000 - val_loss: 0.0655 - val_accuracy: 1.0000 - 74ms/epoch - 5ms/step\n", "Epoch 352/400\n", "16/16 - 0s - loss: 0.0573 - accuracy: 1.0000 - val_loss: 0.0649 - val_accuracy: 1.0000 - 76ms/epoch - 5ms/step\n", "Epoch 353/400\n", "16/16 - 0s - loss: 0.0567 - accuracy: 1.0000 - val_loss: 0.0643 - val_accuracy: 1.0000 - 81ms/epoch - 5ms/step\n", "Epoch 354/400\n", "16/16 - 0s - loss: 0.0561 - accuracy: 1.0000 - val_loss: 0.0637 - val_accuracy: 1.0000 - 115ms/epoch - 7ms/step\n", "Epoch 355/400\n", "16/16 - 0s - loss: 0.0555 - accuracy: 1.0000 - val_loss: 0.0632 - val_accuracy: 1.0000 - 128ms/epoch - 8ms/step\n", "Epoch 356/400\n", "16/16 - 0s - loss: 0.0549 - accuracy: 1.0000 - val_loss: 0.0626 - val_accuracy: 1.0000 - 119ms/epoch - 7ms/step\n", "Epoch 357/400\n", "16/16 - 0s - loss: 0.0544 - accuracy: 1.0000 - val_loss: 0.0620 - val_accuracy: 1.0000 - 75ms/epoch - 5ms/step\n", "Epoch 358/400\n", "16/16 - 0s - loss: 0.0538 - accuracy: 1.0000 - val_loss: 0.0615 - val_accuracy: 1.0000 - 117ms/epoch - 7ms/step\n", "Epoch 359/400\n", "16/16 - 0s - loss: 0.0532 - accuracy: 1.0000 - val_loss: 0.0610 - val_accuracy: 1.0000 - 112ms/epoch - 7ms/step\n", "Epoch 360/400\n", "16/16 - 0s - loss: 0.0527 - accuracy: 1.0000 - val_loss: 0.0604 - val_accuracy: 1.0000 - 112ms/epoch - 7ms/step\n", "Epoch 361/400\n", "16/16 - 0s - loss: 0.0522 - accuracy: 1.0000 - val_loss: 0.0599 - val_accuracy: 1.0000 - 84ms/epoch - 5ms/step\n", "Epoch 362/400\n", "16/16 - 0s - loss: 0.0516 - accuracy: 1.0000 - val_loss: 0.0594 - val_accuracy: 1.0000 - 77ms/epoch - 5ms/step\n", "Epoch 363/400\n", "16/16 - 0s - loss: 0.0511 - accuracy: 1.0000 - val_loss: 0.0588 - val_accuracy: 1.0000 - 78ms/epoch - 5ms/step\n", "Epoch 364/400\n", "16/16 - 0s - loss: 0.0506 - accuracy: 1.0000 - val_loss: 0.0584 - val_accuracy: 1.0000 - 80ms/epoch - 5ms/step\n", "Epoch 365/400\n", "16/16 - 0s - loss: 0.0501 - accuracy: 1.0000 - val_loss: 0.0579 - val_accuracy: 1.0000 - 78ms/epoch - 5ms/step\n", "Epoch 366/400\n", "16/16 - 0s - loss: 0.0496 - accuracy: 1.0000 - val_loss: 0.0573 - val_accuracy: 1.0000 - 119ms/epoch - 7ms/step\n", "Epoch 367/400\n", "16/16 - 0s - loss: 0.0491 - accuracy: 1.0000 - val_loss: 0.0568 - val_accuracy: 1.0000 - 127ms/epoch - 8ms/step\n", "Epoch 368/400\n", "16/16 - 0s - loss: 0.0486 - accuracy: 1.0000 - val_loss: 0.0563 - val_accuracy: 1.0000 - 73ms/epoch - 5ms/step\n", "Epoch 369/400\n", "16/16 - 0s - loss: 0.0481 - accuracy: 1.0000 - val_loss: 0.0559 - val_accuracy: 1.0000 - 124ms/epoch - 8ms/step\n", "Epoch 370/400\n", "16/16 - 0s - loss: 0.0476 - accuracy: 1.0000 - val_loss: 0.0554 - val_accuracy: 1.0000 - 79ms/epoch - 5ms/step\n", "Epoch 371/400\n", "16/16 - 0s - loss: 0.0472 - accuracy: 1.0000 - val_loss: 0.0549 - val_accuracy: 1.0000 - 74ms/epoch - 5ms/step\n", "Epoch 372/400\n", "16/16 - 0s - loss: 0.0467 - accuracy: 1.0000 - val_loss: 0.0545 - val_accuracy: 1.0000 - 114ms/epoch - 7ms/step\n", "Epoch 373/400\n", "16/16 - 0s - loss: 0.0463 - accuracy: 1.0000 - val_loss: 0.0541 - val_accuracy: 1.0000 - 72ms/epoch - 5ms/step\n", "Epoch 374/400\n", "16/16 - 0s - loss: 0.0458 - accuracy: 1.0000 - val_loss: 0.0536 - val_accuracy: 1.0000 - 77ms/epoch - 5ms/step\n", "Epoch 375/400\n", "16/16 - 0s - loss: 0.0454 - accuracy: 1.0000 - val_loss: 0.0531 - val_accuracy: 1.0000 - 115ms/epoch - 7ms/step\n", "Epoch 376/400\n", "16/16 - 0s - loss: 0.0449 - accuracy: 1.0000 - val_loss: 0.0527 - val_accuracy: 1.0000 - 76ms/epoch - 5ms/step\n", "Epoch 377/400\n", "16/16 - 0s - loss: 0.0445 - accuracy: 1.0000 - val_loss: 0.0523 - val_accuracy: 1.0000 - 80ms/epoch - 5ms/step\n", "Epoch 378/400\n", "16/16 - 0s - loss: 0.0441 - accuracy: 1.0000 - val_loss: 0.0519 - val_accuracy: 1.0000 - 73ms/epoch - 5ms/step\n", "Epoch 379/400\n", "16/16 - 0s - loss: 0.0437 - accuracy: 1.0000 - val_loss: 0.0515 - val_accuracy: 1.0000 - 79ms/epoch - 5ms/step\n", "Epoch 380/400\n", "16/16 - 0s - loss: 0.0432 - accuracy: 1.0000 - val_loss: 0.0511 - val_accuracy: 1.0000 - 118ms/epoch - 7ms/step\n", "Epoch 381/400\n", "16/16 - 0s - loss: 0.0429 - accuracy: 1.0000 - val_loss: 0.0507 - val_accuracy: 1.0000 - 113ms/epoch - 7ms/step\n", "Epoch 382/400\n", "16/16 - 0s - loss: 0.0425 - accuracy: 1.0000 - val_loss: 0.0503 - val_accuracy: 1.0000 - 118ms/epoch - 7ms/step\n", "Epoch 383/400\n", "16/16 - 0s - loss: 0.0421 - accuracy: 1.0000 - val_loss: 0.0499 - val_accuracy: 1.0000 - 76ms/epoch - 5ms/step\n", "Epoch 384/400\n", "16/16 - 0s - loss: 0.0417 - accuracy: 1.0000 - val_loss: 0.0495 - val_accuracy: 1.0000 - 90ms/epoch - 6ms/step\n", "Epoch 385/400\n", "16/16 - 0s - loss: 0.0413 - accuracy: 1.0000 - val_loss: 0.0491 - val_accuracy: 1.0000 - 115ms/epoch - 7ms/step\n", "Epoch 386/400\n", "16/16 - 0s - loss: 0.0409 - accuracy: 1.0000 - val_loss: 0.0487 - val_accuracy: 1.0000 - 116ms/epoch - 7ms/step\n", "Epoch 387/400\n", "16/16 - 0s - loss: 0.0405 - accuracy: 1.0000 - val_loss: 0.0483 - val_accuracy: 1.0000 - 116ms/epoch - 7ms/step\n", "Epoch 388/400\n", "16/16 - 0s - loss: 0.0402 - accuracy: 1.0000 - val_loss: 0.0479 - val_accuracy: 1.0000 - 116ms/epoch - 7ms/step\n", "Epoch 389/400\n", "16/16 - 0s - loss: 0.0398 - accuracy: 1.0000 - val_loss: 0.0476 - val_accuracy: 1.0000 - 86ms/epoch - 5ms/step\n", "Epoch 390/400\n", "16/16 - 0s - loss: 0.0395 - accuracy: 1.0000 - val_loss: 0.0472 - val_accuracy: 1.0000 - 142ms/epoch - 9ms/step\n", "Epoch 391/400\n", "16/16 - 0s - loss: 0.0391 - accuracy: 1.0000 - val_loss: 0.0468 - val_accuracy: 1.0000 - 140ms/epoch - 9ms/step\n", "Epoch 392/400\n", "16/16 - 0s - loss: 0.0387 - accuracy: 1.0000 - val_loss: 0.0465 - val_accuracy: 1.0000 - 149ms/epoch - 9ms/step\n", "Epoch 393/400\n", "16/16 - 0s - loss: 0.0384 - accuracy: 1.0000 - val_loss: 0.0461 - val_accuracy: 1.0000 - 161ms/epoch - 10ms/step\n", "Epoch 394/400\n", "16/16 - 0s - loss: 0.0381 - accuracy: 1.0000 - val_loss: 0.0458 - val_accuracy: 1.0000 - 110ms/epoch - 7ms/step\n", "Epoch 395/400\n", "16/16 - 0s - loss: 0.0377 - accuracy: 1.0000 - val_loss: 0.0454 - val_accuracy: 1.0000 - 135ms/epoch - 8ms/step\n", "Epoch 396/400\n", "16/16 - 0s - loss: 0.0374 - accuracy: 1.0000 - val_loss: 0.0451 - val_accuracy: 1.0000 - 147ms/epoch - 9ms/step\n", "Epoch 397/400\n", "16/16 - 0s - loss: 0.0371 - accuracy: 1.0000 - val_loss: 0.0448 - val_accuracy: 1.0000 - 104ms/epoch - 6ms/step\n", "Epoch 398/400\n", "16/16 - 0s - loss: 0.0367 - accuracy: 1.0000 - val_loss: 0.0444 - val_accuracy: 1.0000 - 109ms/epoch - 7ms/step\n", "Epoch 399/400\n", "16/16 - 0s - loss: 0.0364 - accuracy: 1.0000 - val_loss: 0.0441 - val_accuracy: 1.0000 - 110ms/epoch - 7ms/step\n", "Epoch 400/400\n", "16/16 - 0s - loss: 0.0361 - accuracy: 1.0000 - val_loss: 0.0438 - val_accuracy: 1.0000 - 141ms/epoch - 9ms/step\n", "第5个弱分类器训练完毕\n" ] } ], "source": [ "#设置参数\n", "\n", "epochs=400 #迭代次数\n", "batch_size=32 #batchsize\n", "number_of_weak_classifiers=5 #基分类器个数\n", "num_of_res_blocks=1 #基分类器的残差块个数\n", "num_hidden_layers_of_res_block=4 #残差块的隐层数\n", "num_neurons_of_hidden_layer=2 #隐层的神经元数\n", "\n", "#训练模型\n", "boosting_model,boosting_models,history=train_ResBoost_model(number_of_weak_classifiers,\n", " num_of_res_blocks,\n", " num_hidden_layers_of_res_block,\n", " num_neurons_of_hidden_layer,\n", " batch_size=batch_size,\n", " epochs=epochs)" ] }, { "cell_type": "markdown", "metadata": { "id": "0ua1P4IyPAqb" }, "source": [ "### **可视化结果**" ] }, { "cell_type": "markdown", "metadata": { "id": "56snPdb_PBN0" }, "source": [ "#### **绘制训练曲线**" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000 }, "id": "zWImdcnY_meX", "outputId": "759b7049-3c57-407c-b2d2-7a58d1307d6d" }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAf0AAAEjCAYAAAA4+NjXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOydd3gWxdbAfycdktBC7x0NgiCIBREsIIoK1msHr1fFgr1d28V2bdgbNsSOXvxEVKwUERUFpAgoAqGFToCQQvr5/pgNvEneNEjeN+X8nmef7O6ZmT27mXfPzJmzM6KqGIZhGIZR8wkJtgKGYRiGYQQGM/qGYRiGUUswo28YhmEYtQQz+oZhGIZRSzCjbxiGYRi1BDP6hmEYhlFLCKjRF5FrRGSriKSKSFwgr30giIiKSOcKKmusiLxXEWUVU/4yERnk7YuIvCUiu0TkNxEZICIrKuva5UVELhaRb0uQDxKRxEDqdCCIyFciMrIYWXuv/oRVwHUmisjDpaQp8ZmVpYwy6FGpdbjQtYJWR8rwLMeLyH0lyIt9b4jIKBGZUxF6Hsj1qxMicreIvFGCvNKfZU2kzEZfRLqISMaB/uhFJBx4GhiiqjGqmuQnjYpImtcoSC3pH24URFW7q+os7/A4YDDQWlX7qeqPqtoteNoVRFXfV9Uh+ccH+5ISkfNF5GcRSReRWRWiZBlQ1VNV9e1AXa82UZXriKqOVtWHDqYMo3RU9b+q+i84+Ea0iESKyJsisk5EUkRkkYicWrEaVw/K8wBfAuYdxLWaAVHAslLSHa6qqw7iOga0A9aqatrBFiQiYaqaUwE6VSY7gWeBQ4ATy5OxmtyfcfAccB0xSkdEQlU1N9h6lEAYsAEYCKwHTgM+FpEeqro2WEqJiACiqnmBumaZevoicgGwG5heSrpIEXlWRDZ527Peua5Avnt5t4jMOBilReQEEfnD5/g7EZnnc/yjiIzw9luKyCcisl1E1ojIDT7p+onILyKyW0Q2i8iLIhJRzDWPE5EN+S70EnTr7umz0xvKuLuYdP8TkS0ikiwis0Wku4/sNBFZ7rVIN4rIbd75xiLyhafvTu8+QzzZWhE5WUSuAN4AjvG8JQ8UdleW8kzGishkEXlPRPYAo7znNF9E9nj39HQx9/SDiJzj7ff3WubDvOOTRGSRt7/PLScis73siz19/+FT3q0iss3731xe3DNX1e9V9WNgUwn/mvwyB4lIoojcKSJbgLdEJERE7hKR1SKSJCIfi0gjL32U9yySvOc+T0SaebJZIpLfEwkVkXEiskNEEoBhha67VkROLvSc3/M5LrY+lAdxLtEd3vUuLiHdlSKyyqtHU0WkpY+s1DosIuEi8qFXjyJqUh0prWwpNFQiIrd7aTaJyD8LlRHnPd89IvIb0KmQ/BCfZ71CRM4vdJ2XRORLce+CX0WkQP4y3MMwEVnoXX+DiIz1kX0pImMKpV8iImeVUbdXRGSaiKQBJ0gx7y0/Oq0TkT7e/sVeHejuHV8hIlO8fd/fSH4d2O3VgWN8yhsnbihzjRTTe1fVNFUdq6prVTVPVb8A1gB9itFxlIjMKa5sEakvznOw2bvXh0Uk1I/eRbwU4t4bj4jIT0A60FFEjhX3bkn2/h7rk3+WiDwkIj95z/ZbEWnsyYp9PxVHqUZfROoBDwK3lJYWuAc4GugFHA70A+5V1b+B/JdYA1UtqaU9W9zL7/9EpH0xaeYCXcQZwXCgJ9BSRGJFpA7QF8g3iJ8Di4FWwEnATSJyildOLnAz0Bg4xpNf6+cZDAU+BM7xcaEXQURige+Br4GWQGeKbyh9BXQBmgK/A+/7yN4ErlbVWOAwIL+RdCuQCDTBeU7uBgrMo6yqbwKjgV+8YZT/FNKxtGcCMByYDDTw9HoOeE5V6+FeWh8Xc08/AIO8/YFAAnC8z/EPhTOoar78cE/fj7zj5kB9T8crgJdEpGEx1y0vzYFGOI/IVcAYYISnY0tgF86zBTDS06MNEId7tnv9lHklcDrQG1f/zi2nTiXVh7LSHFeXW3l6vyYiRYZ1RORE4FHgfKAFsA6Y5MlKrcPeb2wKkAmcr6pZ1Mw6UmrZ3rvhNtxwWhfg5EJJXgIycM/5n96Wnzca+A74APd/vwB4WUTiffJfADwANARWAY+U8z7SgMtwv+VhwDXidYiAt4FLfPQ53LvfL8uo20WePrHAHIp/bxWm3HXAR97AqwO/eMdH4TqUjYEngDdFRIp7GD732gzoSsme55LKngjk4H4fvYEhwL9Ku64Pl+LePbFACvAl8DzuHfM07n/gG/d2EXA57n8RgatzUPb30z7K0tN/CHhTVcsSNHMx8KCqblPV7bjKemkZ8uUzEGiPc8FtAr4QP2M4qroXN9RwPK6lthj4CeiPa3Ss9GIGjgSaqOqDqpqlqgnA67gKjKouUNW5qprjuXhe9XTw5Tzv/Kmq+lsp+p8ObFHVp1Q1Q1VTVPVXfwlVdYInzwTGAoeLSH1PnA3Ei0g9Vd2lqr/7nG8BtFPVbG+svryLJ5T4TDx+UdUpXot4r3fdziLSWFVTVXVuMWX/wP7ndzzOsOQfF/djLo5sXF3KVtVpQCpQUXEJecB/VDXTu7/RwD2qmujz/zjXq3vZuB9TZ1XN9erMHj9lng88q6obVHUn7t7LTCn1oTzc593XD7gXyfl+0lwMTFDV373r/RvnGWpP6XW4Hq5BsBq43MelW9PqSFnLPh94S1WXesNpY/MFXs/vHOB+r6e5FGdo8zkdNwz3lvcOWgh8gnvn5POpqv7mDUG9j+tQlRlVnaWqf3i/5SW4zkv+854KdBWRLt7xpcBHXiOuLLp9pqo/eWVnUPx7qzC+dWAAB1cH1qnq6149fBv3fiyxp+t1FN8H3lbVv8pbttdgOA24yfu/bgOeoeA7tDQmquoy7/86BGez3vWe9YfAX8AZPunfUtW/vffVx+yvB2V9P+2jRKMvIr1wLddnyngjLXG9hnzWeefKhKrO9gzRbuBGoANwaDHJ81uLx3v7s3AVxrfStMN5AHbnb7jecb57tqs4d/kWca7s/+Jadb7cBHzs/WBLow3uZVgi4lzBj4lzJ+8B1nqi/Gufg6tU68S5Q/NdWU/iWvvfikiCiNxVBp0KU+Iz8dhQKM8VuFbxX5776PRiyv4F9xJphquU7wBtPFdUP/a76MpCkhYca08HYsqRvyS2ey+pfNoBn/o8jz9xXqBmwLvAN8Akce7bJ7yXRmFaUvC5rfOTxi9lqA9lZZcWjOMo7vdX4HeqqqlAEq6XV1odPhrnWXusUIOzptWRspZd0v+9CfvHkv3J2wFHFfotXozzMuSzpQw6FIuIHCUiM8UN5SXjGriNAbzfwEfAJZ4H8EJcfS+rboXfE8W9twrzAzBARFoAoTgj1t9rdNYHFpXjFvc9H1VN93aLfUbefb4LZAHXH2DZ7YBwYLPPs3kV1wsvK77PrrDdxDtu5U8XCtaDsr6f9lFaT38Qrue9Xtz4523AOSJSXAtuE+6B5NOWcoyh+UGB4lw1hY1+fuvR1+hvANaoagOfLVZVT/Pkr+BaVF3UuSXv9nO984ARInJjGfTdAHQsQ7qLcC70k3GVvL13XgBUdZ6qDsdVoil4rlKv13WrqnYEzgRuEZGTynC9wjqW9Eyg6JDBSlW90NPncWCy5/6jULp0YAGuwbbU6zH8jBsaWq2qO8qpa2VR2DuyAefJ8X0mUaq60evpPaCq8cCxuB7QZX7K3IwzmPm0LSRPA+r6HPu+PEusD+WgYaH/S3G/vwK/Uy9PHLCR0uvwt7ie2XTfscMaWEfKSkn/9+04F3Bx8g3AD4XqXYyqXlOB+n2A69G3UdX6wHgK1qu3ccb8JCBd97vNy6Jb4feE3/dWYdQFaqfjhtVmez3TLTh39xz1H9R20MvBeq75N3GN+XNUNfsAi9qAG9pq7PNs6qlq/hB2Sb/1fHzvp7DdBFdPNpamSDneT/sozei/hhuf6+Vt43Euw1OKSf8hcK+INPFa7vcDZfrET1zwUC+v1xMDPIW76T+LyfIzzt3WD/hNVZfhtU7Z31v4DUgRF7RVxyv7MBE50pPHAnuAVBE5BPD3Y9uE+0HcKCKl/Ri/AFqIyE3iAhhjReQoP+licZUmCVc5/uvzHCLEBbfU9yrlHpw7GhE5XUQ6e5U3GdcbLW/UZ2nPpAgicomINPF+jLu908Vd9wdcCzq/4TWr0LE/tlK2xlJx+oWKSBSuVxUiLrilxNZuIcYDj4hIO6+8JiIy3Ns/QUR6eK7aPTh3mr97/xi4QURaixv7LeyFWQRcIC4ArvCYf7H14QB4wKtDA3AvgP/5SfMhcLn3e4v0rveruiGuUuuwqj6BMybTZX9AUU2vI8XxMS7YNV5E6gL7Ymg8t/D/AWNFpK648XDfeR2+wHk9LvXqRbiIHCkixXk3D4RYYKeqZohIP1wDcx+ekc/DvW/f9RGVS7eS3lvFUN46sN0r74DrAK6TdyhwhucmPyBUdTOu8fuUiNQTFwjcSUTyhygWAceLSFtxQ3T/LqXIabhnfZGIhIkLVI3H/Q9KpBzvp32UaPRVNV1Vt+RvuHGtDHXj9f54GJgPLAH+wAUklXVCkGY4V9MeXGBHe+D04lpjnhvzd2CZ11sA5zpc542x5P/oTsc1WNYAO3CR7fljpbfhfgQpuHHtj/CDqq7HGf67xIvWLiZdCi6g5wxcy3UlcIKfpO/g3DcbgeW4wERfLgXWinP1jsa1xMEFCn2P+z/8ArysqjOL06cYHUt7Jv4YCiwTkVRcwNYFJfxofsC9aGYXc+yPscDbnqvM3xh0aVyKC155BTdGuBf3/ywrz+F6Q9+KSAru/5Fv6Jrjghr34BqgP1Dw5ZjP6zg322Jcvfy/QvL7cA3oXbhYlw98ZKXVh7KyxSt/E27McrS/MUtV/d7T5xNcT7UT++NcylSH1X2nPgX4XtyXDjW9jvhFVb/CfQo4Azf0Vjh47XqcK3YLLvjrLZ+8Kbjx3Atw/7MtOC9J5MHq5cO1wINevb4f/73vd4Ae+HTQDlC34t5b/ihXHfA8RI8AP3l14OgSyi6C16C/Gvfe2yL754IpSceSuAwXULcc95ubjBvzR1W/w9mSJTivVonGW1382em4QO0k4A6c7SuL16us76d9iJY7DswwDMOoKYjIZcBVqnpcsHUxKh+be98wDKOW4g1JXIsbyjVqAWb0y4m4eexT/W3B1s2o2YibeMdf3fsq2LoZgaOi3kHi5ubYjouX+KCU5EYNwdz7hmEYhlFLsJ6+YRiGYdQSzOgbhmEYRi3BjL5hGIZh1BLM6BuGYRhGLcGMvmEYhmHUEszoG4ZhGEYtwYy+YRiGYdQSzOgbhmEYRi3BjL5hGIZh1BLM6BuGYRhGLcGMvmEYhmHUEszoG4ZhGEYtwYy+YRiGUSMQkaEiskJEVonIXX7ko0XkDxFZJCJzRCTeO99eRPZ65xeJyPjAax8YbJU9wzAMo9ojIqHA38BgIBGYB1yoqst90tRT1T3e/pnAtao6VETaA1+o6mEBVzzAhAVbgdpISEiI1qlTJ9hqGIZhVCvS09NVVYvzUPcDVqlqAoCITAKGA/uMfr7B94gGal2v14x+EKhTpw5paWnBVsMwDKNaISLZIjLf59Rrqvqat98K2OAjSwSO8lPGdcAtQARwoo+og4gsBPYA96rqjxWqfBXBjL5hGIZRXchR1b4HU4CqvgS8JCIXAfcCI4HNQFtVTRKRPsAUEeleyDNQI7BAPsMwDKMmsBFo43Pc2jtXHJOAEQCqmqmqSd7+AmA10LWS9AwqZvQNwzCMmsA8oIuIdBCRCOACYKpvAhHp4nM4DFjpnW/iBQIiIh2BLkBCQLQOMObeNwyj1pOdnU1iYiIZGRnBVsUAoqKiaN26NeHh4WXOo6o5InI98A0QCkxQ1WUi8iAwX1WnAteLyMlANrAL59oHOB54UESygTxgtKrurMBbqjLYJ3tBIDo6Wi2QzzCqDmvWrCE2Npa4uDhEJNjq1GpUlaSkJFJSUujQoUMBmYikq2p0kFSrEdQq976ITBCRbSKytBi5iMjz3sQOS0TkCB/ZSBFZ6W0jfc738SZ7WOXltTeGYVQzMjIyzOBXEUSEuLg487pUErXK6AMTgaElyE/FjeV0Aa4CXgEQkUbAf3Cff/QD/iMiDb08rwBX+uQrqXzDMKooZvCrDva/qDxq1Zi+qs72Zl4qjuHAO+rGPOaKSAMRaQEMAr7LH+MRke+AoSIyC6inqnO98+/gokG/qgz9/2/l/xEiIYzoPKIyijcMv6gq785dx46UzGCrUmn0b5zNluTy9ywFpW72TkTzisjSIhqjCJG5KYTnFi07NaIJAFE5ewjLK/hsVYS08MaePJmwvKxC8hDSwuMAqJOzm9C87ALyPAklPbwRAHWzdxGiOQXlIWGkhzX05DsJ0dwC8tyQcPaGNQAgOjupyP3lhESSEVbPk+9ACg0T+8pjsrbvO58RFkvD+vUJDTGjHixqldEvA/4md2hVyvlEP+eLICJX4bwHREREHJByX676AklNN6NvBJT563Zx/2fLAKipHbCeZ7RgW8qBGH1oKHm0ZAeFH82azFgUaCl7aEDBz70VISEzFoA2kkzGrkRO/sdoALZsTyI0JITYuGYA/PjFB9SPKGi0swhjTYYb2t61ZA6TJ3/C8w/dsU+eQQTbMuoC0El2UZdM+g8fxU+fTQQgnUi2qZsVtKvsJJKCjYZU6rBNowA4RHYSzv7rz/p5Po+9+gFPTvwUgHhJIpSCjYJdxLJN3Xuuqex/NjuzQqhfrx6hRZ6WESjM6AcIb9ao18AF8h1IGRe8sx7dvBXOq1DVDKNY0rNyeGHGKiLCQlh4y+FE791cNFGzwyAsEvZshpRNReXNe0JoOCRvhNQtReUtekNICOxeD2nbCwkFWnmhNbvWQnpSIXEotOzl9pNWQ8bugvLQCGjew+3vWAmZheZaCasDzeL5888/ObRZFBTq8SIhEO5NmZ29Fwr36CUUwhtQ8PNwR499ew2KyATo6Stv1YVFy1YAMHbsWGJiYrjtttv2pcjJySEsbP/rOtI3f+tzGXjauQXKr1O4fODn+Yv3nYn2I/cltgS5NEkmPCqanq0b+JUDNPI2R+99e22LpDQCjRn9ghQ3ucNGnIvf9/ws73xrP+krhZAWzWi0dCOpWanERMRU1mUMYx93f7KE2X9vY2j3FkSv/AK+ur1oohsXQ8P2sPgDmP5gUfntCRAdB/PegDlPF5Xfuw1CIuHnF+C31wrKQsLh/h1u/4cnYNH7BeV1GsKda93+92Phz6kF5fXbwM1e3O5Xd8DqGQXlTQ6F6+a6/d3rIbvQVzXhdaFJN7e/ay3kFPIGRMRC485F76kCGDVqFFFRUSxcuJD+/ftzwQUXcOONN5KRkUGdOnV466236NatG7NmzWLcuHF88cUXjB07lvXr15OQkMD69eu56aabuOGGGwCIiYkhNTWVWbNmMXbsWBo3bszSpUvp06cP7733HiLCtGnTuOWWW4iOjqZ///4kJCTwxRdflEnfDz/8kP/+97+oKsOGDePxxx8nNzeXK664gvnz5yMi/POf/+Tmm2/m+eefZ/z48YSFhREfH8+kSZMq5RkaRTGjX5D87zgn4YL2klV1s4h8A/zXJ3hvCPBvVd0pIntE5GjgV+Ay4IXKUi6qfQeisn9nw7o/OLTLMZV1GcMAYG9WLluX/8ic6Neod9x70OAUaNiuaMLopu5v/AjX6y9MpHNjc/iF0PboovIQ7zXU53LofHIhoY8b+OhrIH64/7wAA26B3pcUlIdF7d8/4R44anRBeYTP11/1WoLm8vjvz/HX7pX7rx8S6nbzcimyPouI6+2XwiGNDuHOfneWmq4wiYmJ/Pzzz4SGhrJnzx5+/PFHwsLC+P7777n77rv55JNPiuT566+/mDlzJikpKXTr1o1rrrmmyPfuCxcuZNmyZbRs2ZL+/fvz008/0bdvX66++mpmz55Nhw4duPDCC8us56ZNm7jzzjtZsGABDRs2ZMiQIUyZMoU2bdqwceNGli51Da/du50n5rHHHmPNmjVERkbuO2cEhlpl9EXkQ1yPvbGIJOIi8sMBVHU8MA04DVgFpAOXe7KdIvIQbsYngAd9Jm64FvdVQB1cAF+lBPEBNOocD3zCthWLzejXID5ZkMijX/1FSQt+hWgueZ5xuSb3Q0bo9wXkijA07A0Absl9i1N0TgF5GnU4O+xFAO7JfYXjdX4B+Q4acnHYOAAeyX2Gvupe0q+HZFJX8whp0Qmi6vk3+vnEdXJbcTTp6rbiaBbvtuJo3mO/q94fLXsXLwNoXcqU7ZGe9ywssmBjIp+Q0o17RXPeeecRGuqum5yczMiRI1m5ciUiQnZ2tt88w4YNIzIyksjISJo2bcrWrVtp3bp1gTT9+vXbd65Xr16sXbuWmJgYOnbsuO/b+AsvvJDXXnutSPn+mDdvHoMGDaJJExecePHFFzN79mzuu+8+EhISGDNmDMOGDWPIkCEA9OzZk4svvpgRI0YwYoTFKAWSWmX0VbXEpqsXtX9dMbIJwAQ/5+cDAVmDuUW3I9gM7ElYEYjLGQHi7V/WEhkWwqBuTYpNc+76h4jKTeO9Do9Rd/fhrErNLZLmlNbNAQjf2ZtV6QV7dtkSySmtnJykPqzaG1tAvjc0llNaOHnWjn6syvAiy8ND6dHvRGfwawkH0iOvLKKj93si7rvvPk444QQ+/fRT1q5dy6BBg/zmiYyM3LcfGhpKTk7OAaWpCBo2bMjixYv55ptvGD9+PB9//DETJkzgyy+/ZPbs2Xz++ec88sgj/PHHHwViFozKw55yNaJ+u868Nqgukc1sFsUDJSsnj6e+XcHudP+9pECTq8qSxGT+feohXD2wUC/5+wf2B7al/QTdz+KR4T3wDRHzZf8aov7lx5UiH1SK3AguycnJtGrlPg6aOHFihZffrVs3EhISWLt2Le3bt+ejjz4qc95+/fpxww03sGPHDho2bMiHH37ImDFj2LFjBxEREZxzzjl069aNSy65hLy8PDZs2MAJJ5zAcccdx6RJk0hNTaVBg6IBgUbFY0a/GiFhYSw+4xDqhKUGW5Vqy8wV23h1dgKNYyIJqyLfCnduGsPwXn6+9Fz3swsuA6gbB31GBVQvo2pxxx13MHLkSB5++GGGDRtW4eXXqVOHl19+maFDhxIdHc2RRx5ZbNrp06cXGDL43//+x2OPPcYJJ5ywL5Bv+PDhLF68mMsvv5y8PPfVw6OPPkpubi6XXHIJycnJqCo33HCDGfwAYnPvB4GDmXv//mk3k7h6ERPGzKxgrWo+2/ZkcNEbv7IjNZN595xMeGgAJ6TMyYTVMyHPc6M2PdSNf2emQsKs/elSNrvjUx5xEfFGQPjzzz859NBDg61G0ElNTSUmJgZV5brrrqNLly7cfPPNQdHF3//E5t4/eKynX8047ttNxM3YQtY1WUSEHdgkP7WVWz5ezKptqVx8VNvAGnyAX1+F7+7bfzz4Ieh/A6RuhY8uLpg2vC5EWc/HCDyvv/46b7/9NllZWfTu3Zurr7462CoZFYwZ/WpGZLv21M1awsYNy+nQoVew1anyqCq5ecrOtCx+Xr2DYT1bcP8ZJUSIVxbL/s9NUjP8JXcc28L9rd8arv6xYNroxlDHjL4ReG6++eag9eyNwGBGv5rRoNMhwFQ2r/jdjH4ZuH3yEiYv2D9T8nWDOhMZFuBPr3Kzoe2xENsMWvQsKAuLLHrOMAyjkjCjX81oHt+HncCev5fben6lkJKRzdTFm+jfOY6jO8TRrH4U8S0D+OlZ9l746TnodREcfoGN0RuGEXTM6FczmnTozuZwyFm9JtiqVEl+Xr2D12YnoArJe7PJysnjlsFd6dOuUemZy8OKr2He624/uimc9Yrbn/4gbPbmOM/YA4m/QZt+0OnEir2+YRjGAWBGv5oREhrK5AtbE9ImltODrUwV5MUZq/gjMZmOTd3samcc3pLebRqWkusAmPVf9zldo44Qun+iEzJTYO+u/ceHngntj6/46xuGYRwAZvSrIanHdCchOSHYalQpFq7fxcL1u5mbkMS1gzpz2yndylfAxgWuZ97pBHc8782iK6o1PRTaH+cWXtm8GIY8DMeOKZjmtCcP+B6M2ktSUhInnXQSAFu2bCE0NHTflLa//fZbqctxz5o1i4iICI499tgisokTJzJ//nxefPHFilfcqHaY0a+GdMqLI3vuTLJP2kN4bO2ZHrU4cnLzuPKd+exIzSIsRBjeq2X5CsjLg0kXu9XU8o3+1/+G3MyC6Y78lzP6sS3cQi7xNme4UTHExcWxaNEiwP/SuqUxa9YsYmJi/Bp9w/DFjH41pNvWUE6eksXGYT/TfkDVjebbkZpJZk5e6QkPksUbdrMjNYunzjucUw5rTkxkCdU6y5sUKSIasjPcNLdblrhJcQbdtT/dLcuL5g3z3PihEXBHQsEV2gyjglmwYAG33HILqampNG7cmIkTJ9KiRYsiy9I+9thjjB8/ntDQUN577z1eeOEFBgwYUGr5Tz/9NBMmuOVE/vWvf3HTTTeRlpbG+eefT2JiIrm5udx333384x//4K677mLq1KmEhYUxZMgQxo0bV9m3b1QSZvSrIc26Hwm8y+alv1VZo79g3U7OeeWXgF2vbkQop/VoQZ2IEj7Hy8uFl4+GE++HnufBhl/hnTOdLKwOHHbu/rTRjYsvR8QMfg1n3aWXFTkXe+pQGl10EXl797LhqqKT1tQ/6ywanH0WObt2sfGGGwvI2r37Trmur6qMGTOGzz77jCZNmvDRRx9xzz33MGHChCLL0jZo0IDRo0eXyzuwYMEC3nrrLX799VdUlaOOOoqBAweSkJBAy5Yt+fLLLwE3339SUhKffvopf/31FyJiS+FWc8zoV0M6djuKVeGQtvKvYKtSLAvXuxfDQ8O7B+S7+M7NYko2+ADr57rgu/xAu8Zd4cwX9+/nL61qGEEmMzOTpUuXMnjwYAByc3Np0cJN6FQRy9LOmTOHs846a98qfmeffTY//vgjQ4cO5dZbb+XOO+/k9NNPZ8CAAeTk5BAVFcUVV1zB6aefzumnWwhxdcaMft8nXZwAACAASURBVDUkNqoe25qGE7ZmQ7BV8ctT367ghRmraBwTwaXHtA+2OvtZPsWNxfe6yB3XawFHXBpcnYwqSUk985A6dUqUhzVsWO6efWFUle7du/PLL0W9Zf6Wpa0ounbtyu+//860adO49957Oemkk7j//vv57bffmD59OpMnT+bFF19kxowZFXbNikREhgLPAaHAG6r6WCH5aNzy6blAKnCVqi73ZP8GrvBkN6jqN4HUPVAEeAJyo6JIbd2ImMSq52bbnZ7FK7NWA9CxSRXqOeflwfKp0GWw9eiNKk9kZCTbt2/fZ/Szs7NZtmxZgWVpH3/8cZKTk0lNTSU2NpaUlJQylz9gwACmTJlCeno6aWlpfPrppwwYMIBNmzZRt25dLrnkEm6//XZ+//13UlNTSU5O5rTTTuOZZ55h8eLFlXXbB4WIhAIvAacC8cCFIlJ4zu0PVLWHqvYCngCe9vLGAxcA3XHTnr3slVfjsJ5+NWXbPwYxPmEKX+XlEBZSNf6NyenZXPnOfHLydN9xUFg9E9Z689nHdXY9++1/QvoOi7g3qgUhISFMnjyZG264geTkZHJycrjpppvo2rWr32VpzzjjDM4991w+++wzv4F8EydOZMqUKfuO586dy6hRo+jXrx/gAvl69+7NN998w+23305ISAjh4eG88sorpKSkMHz4cDIyMlBVnn766YA+i3LQD1ilqgkAIjIJGA7si8pV1T0+6aOB/GVmhwOTVDUTWCMiq7zyAheYFCBsad0gcDBL6+bz2arPuPene5k6Yiod6neoIM0OjpdmruLJb1ZwaIt6JKdn8cDwwxgc3yywSuTlwtOHQuo2CAmFzoPhoklO9uYpcMkn1tM3imBL61Y9illaNwvwHc94TVVf82TnAkNV9V/e8aXAUap6faEyrgNuASKAE1V1pYi8CMxV1fe8NG8CX6nq5Mq5u+BRNbqIRrnpXLcdZ/yax/pGX9HhjGuDqktaZg5/bUnh88Wb6N22AZ9e27/yLpad4T6xK0yDdm5Bm9Uz3XK157wJPc4tmOaKGjlEZxi1iRxV7XswBajqS8BLInIRcC8wskI0qyaY0a+mtI/rzLlz8kgK/wGCbPTv/vQPPlu0CYD7Tq/kZWtTNsObg4ueP20c9LvS9eLDo6Fr1fyU0TCMSmMj0MbnuLV3rjgmAa8cYN5qS60y+mWI7GwHTACaADuBS1Q1UUROAJ7xSXoIcIGqThGRicBAINmTjVLVRZV7JxAdGcPmFpFErwpuBP/erFy+XbaVUw9rzqVHt6Nv+wpe2CafDy6ArqdAz384F31hGnvT7jaNh6tmmQvfKDeqiogEWw0D9784AOYBXUSkA85gXwBc5JtARLqo6krvcBiQvz8V+EBEngZaAl2A3w5EiapOrTH6PpGdg4FEYJ6ITM3/XMNjHPCOqr4tIicCjwKXqupMoJdXTiNgFfCtT77bgzH2k9KxKa3nJKK5uUhoYAJNN+7ey/njfyEtK4e+7RoyZ9UOMrLzuPTodhzb2WdCmxVfwZRrihZw8SfQug/8MRmm+ZlI5PKvoekhsOBt+P4/7pwqZOyGjoMgoi50Prl4BaPquc0wykFUVBRJSUnExcWZ4Q8yqkpSUhJRUVHlzZcjItcD3+A6dhNUdZmIPAjMV9WpwPUicjKQDezCc+176T7GBf3lANepam7F3VXVodYYfcoQ2Yn7zOMWb38mMIWinIsL8EivRF3LRMghXYiYuYG0VX8T0y0wQUhTFm5k4+699GvfiO//3AbAhf3aclTHOGecF7wFXYZAvVbQ47yiBdT1PAEN2vmXR9V3f+M6FZSHRUGvCyv4bgzD0bp1axITE9m+fXuwVTFwjbDWrVuXO5+qTgOmFTp3v8/+jUUy7Zc9AjxS7otWM2qT0W8F+PrCE4GjCqVZDJyNGwI4C4gVkThVTfJJcwHet50+PCIi9wPTgbu8zz4KICJXAVcBpa6YVVbqH34EuTKDjX/Np1uAjP60PzbTu20DHj7rMIY8M5sbGv7MLfIFfIZboGbpJzDsaTjyCmhRwopzbY50W3G0P85thhEAwsPD6dChanwFYxiViU3OU5DbgIEishA3Tr8RNzsTACLSAuiBcx/l82/cGP+RQCPgTn8Fq+prqtpXVfuGhVVMW6vdYcdy2a2hrOnZpELKK411SWks27SHu2On0XXN+4zo1ZJhLdNg7Ry3bZgHzXvYt/CGYRhVlNrU0y81OlNVN+F6+ohIDHCOqvpOe3c+8KmqZvvk2eztZorIW7iGQ0Do0LAjuRGhrNq9qlLKT07P5ufVO/bNXjH77+2EkUOfje9D3cE8e8FooHelXNswDMOoeGqT0S9LZGdjYKeq5uF68BMKlXGhd943TwtV3Swu+mcEsLSS9C9CZGgkQzc0osuUj9D/XVPhwXz/nfYnH80v+HXAFS3WE7JrF3S33rxhGEZ1o9YY/TJGdg4CHhURBWbjFmYAQETa4zwFPxQq+n0RaQIIsAgYXcm3UoB20oS2y/8ga80aIjt3rrBys3Pz+HrZFoZ2b87Ng7u6k5pHx7nTIC0WOp1UYdcyDMMwAkOtMfpQpsjOyYDfT+9UdS0uGLDw+RMrVsvyUeew7sAf7FmyiCYVaPR/Xp1E8t5szunTmm7NY93JtXNg8Xsuqj68fJ/TGIZhGMHHAvmqOa0PO4rMMNi2sGLXhZi2ZDMxkWEM6NLYrUOfus19ZnfyWDjp/tKyG4ZhGFUQM/rVnPimPVjbDPYurbg1tbNz8/hm+RZOPrQpUQtegwmnwOc3QoM2cNzN0KBthV3LMAzDCBxm9Ks5LaNb8nfnOuysk3egU1cW4ZfVSexOz+b0+IYw/SF38pjrSs5kGIZhVHlq1Zh+TUREWH3ukczL2MmQCpo+dNofm2kYkcsJf42F7DS4dIpNlGMYhlEDsJ5+DSA+Lp5Vu1aRkb33oMvKzs3jm2VbOLdjHqFp2yC6CbQfUAFaGoZhGMHGevo1gPi4eO74MJPV826g+wuvH1RZUxZuZFd6Nn36Hg1dh0BIGIRaNTEMw6gJWE+/BhAfF8/eSMhZdHDBfN8v38rtk5cQERbCoA513Ip2YRWzToBhGIYRfMzo1wBaRrdkQ9u6RGxPJnvbtgMuZ+GGXQB8eOVRRE04ET4N6DxDhmEYRiVjRr8GICLQoxsAexcuOuByVm5NpXPTGPq0ioadCVC/TemZDMMwjGqDGf0aQvMjjiUjHHb/NPuAy1i5LZUuTWMgORE0DxrZUqOGYRg1CTP6NYQeLXrzSf8Qth7a9IDyf7owkTU70ujSLBb2eIsP1isy67BhGIZRjTGjX0Po0bgHU48JZUF8+QPvVJV7PnWLAx7bKQ72bHKC+q0rUkXDMAwjyJjRryHERsTSqUEn1vz1K5kJCeXKuzk5g/SsXB4acRhHd4yDJt3cdLv1WlaStoZhGEYwMKNfg+jZuAfDx/3G9uefL1e+ldtSAdx4PkDL3m5hnfA6FaugYRiGEVTM6NcgDm/ai0XtlZSff0Jzcsqcb+XWFMDH6CcnQsaeylDRMAzDCCJm9GsQfZv1ZUEXgT2ppC/4vcz5/tqSQqPoCOJiIuG31+GZ7vDJvypRU8MwDCMYmNGvQbSJbcPm+GbkhIWQOmN6mfLk5imzVmzjmI5x7kTCLPf3+NsqR0nDMIxKQkSGisgKEVklInf5kd8iIstFZImITBeRdj6yXBFZ5G1TA6t54DCjX4MQEXq3P4Y/O4SRMnNmqekzsnMZ9dZv7EjN4rQeLdzJPZug00nQpl8la2sYhlFxiEgo8BJwKhAPXCgi8YWSLQT6qmpPYDLwhI9sr6r28rYzA6J0EDCjX8M4svmRvHFCLrmvPFxq2m+WbeHHlTvo0DiaEw/xvu/fs9Gi9g3DqI70A1apaoKqZgGTgOG+CVR1pqqme4dzgVr3XXKtMvplcP2081w+S0Rkloi09pH5df2ISAcR+dUr8yMRCeoKNf2a92NznDA/a2WxaVIzc/hyyWbe/3U9TWMjmX7LQOpEhIIqdBkM7Y8LoMaGYRhlJkxE5vtsV/nIWgEbfI4TvXPFcQXwlc9xlFfmXBEZUYE6VylqzZqpPq6fwbjKME9Epqrqcp9k44B3VPVtETkReBS41JPtVdVefop+HHhGVSeJyHhcRXql0m6kFFrGtKR1TGs2f/UZWz5ZS/N77ymS5qWZq3hl1moA/nVcB0JCxAlEYPhLgVTXMAyjPOSoat+DLURELgH6AgN9TrdT1Y0i0hGYISJ/qOrqg71WVaM29fRLdf3gxoFmePsz/cgLICICnIgbGwJ4Gwh6C/H41seTtvIvdr33HtmbNxeQ5eTmMWXhRo7tFMf0Wwdy16mH7BdmZ0BudoC1NQzDqBA2Ar6rhLX2zhVARE4G7gHOVNXM/POqutH7mwDMAnpXprLBojYZ/bK4fhYDZ3v7ZwGxIuKFtft1/cQBu1U1/6P40txJAWFQm0HM6ZoHwJ5p+71XqsrRj85gc3IGI3q1olOTGMJCvSqQsgUeaQYPNXYr7BmGYVQv5gFdvCHXCOACoEAUvoj0Bl7FGfxtPucbikikt98Y6A/4eoFrDLXJ6JeF24CBIrIQ5/bZCOR6snaeW+ki4FkR6VSegkXkqvxxqJxyTJxzIPRt1pfUZrEkdWhE8hdf7Du/My2LHamZHNI8luG9CwXrLZvi/h5/BzS01fUMw6heeJ2v64FvgD+Bj1V1mYg8KCL50fhPAjHA/wrFZx0KzBeRxTgv72OFhn5rDLVmTJ8yuH5UdRNeT19EYoBzVHW3J9vn+hGRWTjXzydAAxEJ8yqcX3eSl+814DWA6OhorbjbKkp4aDj9W/Vn+qGzOX/an2SuXElkly5sTs4A4KaTuxIZFuoSb10G39wDW5dC03g4sWgMgGEYRnVAVacB0wqdu99n/+Ri8v0M9Khc7aoGtamnXxbXT2MRyX8m/wYmeOf9un5UVXGtwnO9PCOBzyr9TsrAwNYD+bZzOtq9C7kpbprdfKPfon7U/oSNOsJh50BcZxh4ZzBUNQzDMAJErTH6ZXT9DAJWiMjfQDPgEe98Sa6fO4FbRGQVboz/zYDcUCkMaDWA1JhQpt87mLpHHAHAluS9QCGjH14HjrgU/vk1dA96DKJhGIZRiYjrrBqBJDo6WtPS0ir9OqO+HkVqViofnfAWOVu38uyqHF6fncDfD5+6/zO9Oc+6VfU6Diy5MMMwjCAjIumqGh1sPaoztaanXxs5sc2JrNi1glWjLmXjnXcxLyGJZvWi9hv8nCyY8RCs+SG4ihqGYRgBwYx+DWZI+yEArDimFZnLlpG+cCFtGtXZn2DbcsjLgSaHBklDwzAMI5CY0a/BNI9uTu+mvXm//UYyo6IZvvpHHj+n5/4Ef34OEgIdBwVLRcMwDCOAmNGv4ZzS/hSWp63mtx59OG7zH7RI3eEEP78AP45z8+zHNAmukoZhGEZAMKNfwxncbjCCMPnwcPJCfZbc7ftPOOZ6OMG+yzcMw6gtWPR+EAhU9H4+o78bzZz1S7ks4hHuuNSi9A3DqJ5Y9P7BYz39WsBp7UcgYckkN9kKQPbnD8Hc8UHWyjAMwwg0ZvRrAR2j+5GXE8Oa9K9J+fAFVt3xPnvnzgy2WoZhGEaAMaNfC1i3I5Oc5CNYvmceqX+OJTQyj60zk9G8vGCrZhiGYQQQM/q1gK+XbaFuxvEo8GGneJrecA17V6xj1wcfBls1wzAMI4CY0a/hZOfmMePPrQw95FAGNz6cyWGZhF1yJdEDBrDtqafIWrcu2CoahmEYAcKMfg1nzY400rJyOapDHCOPupOUnL1MWTWFFg89SEidOuxdvDjYKhqGYRgBwox+DefvrW5Z3UPqZdJj9xZ6N+7Be3++R0jTJnT+7lvqn3lmKSUYhmEYNQUz+jWclVtTCRHouHcZvH8ul7UcxMbUjXy99mtCot3nrikzZ5KZsCbImhqGYRiVjRn9Gs7KbSm0bVSXiLTNAJzY+Uy6NezGiwtfJDs3m9yUFDbf9W823XkneVlZQdbWMAzDqEzM6Ndw1u5Ip0PjaNi2DMKjCYluyo1H3EhiaiKfrPyE0NhYmj/8EBl//MGWBx7AZmg0DMOouVQ7oy8iZ4hItdM7WKRkZtOoTgj8+QV0HQIhIRzX6jj6NOvD+MXjSc9Op97gwcRdM5rkT/6PXe9/EGyVDcMwjEqiOhrPfwArReQJETkk2MpUdVIzcjgkZwWk74D4EQCICDcdcRNJGUm8u/xdAJqMGUPMoEFsffRRstauDaLGhmEYRmVR7Yy+ql4C9AZWAxNF5BcRuUpEYoOsWpVDVUnNzGFXXC8Y8zt0GbJP1qtpL05qexJvLn2TbenbkJAQWo57kpaPPUpE+/bBU9owDMOoNKqd0QdQ1T3AZGAS0AI4C/hdRMYEVbEqRmZOHtm5SnRUJMR1goi6BeS39r2VnLwcnvv9OQBCY2Kof8YZAOxdtozsjRsDrrNhGMaBIiJDRWSFiKwSkbv8yG8RkeUiskREpotIOx/ZSBFZ6W0jA6t54Kh2Rl9EzhSRT4FZQDjQT1VPBQ4Hbi0lb2kVop1XEZaIyCwRae2d7+V5FJZ5sn/45JkoImtEZJG39arI+z0YUjJyADhqy/uw/LMi8jaxbRjZfSRTV09l3pZ5+87nZWWReN31rL/qanJ37w6YvoZhGAeKiIQCLwGnAvHAhSISXyjZQqCvqvbEdRyf8PI2Av4DHAX0A/4jIg0DpXsgqXZGHzgHeEZVe6jqk6q6DUBV04EristUxgoxDnjHqxAPAo9659OBy1S1OzAUeFZEGvjku11Ve3nbogq4xwohY/sargr9nJ6rX4dV0/2muarnVbSNbct9P91HenY6ACEREbR84nGy169nw3XXk5eeHki1DcMwDoR+wCpVTVDVLJwneLhvAlWd6dkKgLlAa2//FOA7Vd2pqruA73Dv+hpHdTT6Y4Hf8g9EpI6ItAdQVf+WzVFqhcA1BmZ4+zPz5ar6t6qu9PY3AduAJgd7I5VN3fkvc3f4h0TkpED74/ymqRNWhwf7P8im1E08s+CZfeej+/Wj5ZNPsHfhQjZcex15GRmBUtswDONAaAVs8DlO9M4VxxXAVweYt9pSHY3+/wDfNWFzvXOlUZZ/6mLgbG//LCBWROJ8E4hIPyACF0iYzyOe2/8ZEYn0d3Ev2HC+iMzPyckpg7oVQOpWVue1YO5Ff0LP84tN1qdZHy4+9GImrZjEb5v3taeoN3QoLR/9L+m//srOt98JhMaGYRglEZb/HvW2qw6kEBG5BOgLPFmx6lV9qqPRD/N66gB4+xEVVPZtwEARWQgMBDbiGhUAiEgL4F3gclXNb3j8GzgEOBJoBNzpr2BVfU1V+6pq37CwsApSt2QSWw7ljdzTiPGm2y2JMb3H0Ca2Dff/fD+pWan7ztcfPpy2b00g7p+XV6aqhmEYZSEn/z3qba/5yDYCbXyOW3vnCiAiJwP3AGeqamZ58tYEqqPR3y4i+1aJEZHhwI4y5Cv1n6qqm1T1bFXtjasUqOpu7zr1gC+Be1R1rk+ezerIBN7CDSNUCVY3HcyHuScRG1V6I6NueF0e7v8wm9M288AvBWfmiz76aCQ8nJwdO9j62ONodnZlqm0YhnEgzAO6iEgHEYkALgCm+iYQkd7AqziDv81H9A0wREQaegF8Q7xzNY7qaPRHA3eLyHoR2YDrWV9dhnxlqRCNfWb7+zcwwTsfAXyKC/KbXChPC++vACOApQd8ZxXNzgSi2UtMZNk8C0c0O4Ixvcfw9dqv+WjFR0XkaT/9xM6JE9l4621ooIYoDMMwyoCq5gDX44z1n8DHqrpMRB706Sg+CcQA//O+tprq5d0JPISzE/OAB71zNQ6prnOti0gMgKqmlpbWJ89pwLNAKDBBVR8RkQeB+ao6VUTOxUXsKzAbuE5VM73xn7eAZT7FjVLVRSIyAxfUJ8AiYHRpOkVHR2taWlqZ7/WAyMtFH4zj+ZyzGD32DSLDQsuWTfO4bvp1/Lr5V9497V26x3UvIE+aOJFtjz1O7ODBtHpqHBJRUSMrhmEYJSMi6apa+nilUSzV0uiLyDCgOxCVf05VHwyeRuUjIEY/dTuM68yDuaO4/6HnypV1V8Yuzvv8PMJCwvj4jI+pF1GvgHzn22+z9dHHiBk4kFbPP0dIpN/YRcMwjAqlKhl9EYkG9qpqnoh0xcV2faWqVXr8s9q590VkPG7+/TG43vV5QLsSM9VG0l2Yg9ZtXO6sDaMaMm7gOLambeXeOfeSp3kF5I1GjqT52LFkbUwkr7IbL4ZhGFWT2UCUiLQCvgUuBSYGVaMyUO2MPnCsql4G7FLVB4BjgK5B1qnKkbpzCwBdOrQ/oPy9mvbi5j43M3PDTF5e9HIRecML/kGH//s/who1QrOzzfgbhlHbEG+in7OBl1X1PJwHukpTHY1+/iwx6SLSEsjGzb9v+LD67z8AODz+wBcivDT+UkZ0HsGrS17ly4Qvi8hDvPH8TXffw/p/XkHunj0HfC3DMIxqhojIMcDFuC+7wMWLVWmqo9H/3JsC90ngd2AtYIvAF2Ieh/FIzsV0PrT3AZchItx/9P30adaH+3+6n0Xb/M8wHDtkMHuXL2fdqFHk7Np1wNczDMOoRtyE+8rrU+8rgY64mVyrNNUqkM/7nO5oVf3ZO44EolQ1ObialY9ABPJd894CVmxJYcZtgw66rF0Zu7joy4tIz0nnw2Ef0jKmZZE0qbNnkzjmBiLatqXtWxMIa1z+WALDMIySqEqBfL54tinGWwG2SlOtevreLHgv+RxnVjeDHxDSdtAs8Rt6Nc4tPW0ZaBjVkJdOeons3GzGzBhDWnbRBkvM8cfT5tXxZCUmsuHa66hOjUnDMIzyIiIfiEg9L4p/KbBcRG4Ptl6lUa2Mvsd0ETnHmwzH8MM3333F2IzH6RNTcXNLdGzQkXEDx7F692pu++E2svOKfpUSffTRtH3jdZrddSf27zEMo4YT7/XsR+AW7umAi+Cv0lRHo381boGdTBHZIyIpIlLlXSqBQlWZu3AJAL26V2wg6bGtjuXeo+9lzsY5PPTLQ35783X79KHuEUcAsGvSR2SuWVOhOhiGYVQRwkUkHGf0p3rf51d5F2dgVn6pQFQ1Ntg6VGW2pWTSIHc7eWEhdO9a8V8yntv1XLamb2X84vE0rduU63tf7zdd7u7dbH/+eRCh7YQ3ierWrcJ1MQzDCCKv4gLJFwOzRaQdUOU7oNWupy8ix/vbgq1XVeHvrSm0JInsuk0htHLadNcefi1ndzmbV5e8yscrPvabJrRBA9q99y4SHs66Sy9j7yL/kf+GYRjVEVV9XlVbqepp3qJr64ATgq1XaVQ7ow/c7rPdB3wOjA2mQlWJlVtTaS47CanfqtKuISLcd/R9DGg1gEd+fYQZ62f4TRfZsSPt3nuP0AYNWPfPK0j75ZdK08kwDCOQiEh9EXlaROZ721NAlfuyoDDVzuir6hk+22DgMMA+DvdYtT2VR0NHEzb8+Uq9TlhIGOMGjiO+UTx3zL6D+Vvm+00X0boV7d57l4jWrclav6FSdTIMwwggE4AU4Hxv24NbmK1KU62+0/eHF8W/TFXjg61LWanM7/RHTviNnWlZfD7muEopvzA7M3Yy6utRbEvfxhtD3uCwxof5TZeXlbVvBr+cpCTC4uICop9hGDWHqvSdvogsUtVepZ2ralS7nr6IvCAiz3vbi8CPuJn5DCB31wYu1c9hz+aAXK9RVCNeH/w6DSIbMPr70azctdJvunyDn7F8OasHD2HXpEkB0c8wDKOS2Csi+3pXItIf2BtEfcpEtTP6wHxggbf9AtypqpcEV6WqQ/OUJZy/czykbQ/YNZtFN+P1Ia8TGRLJVd9dxbo964pNG9GpE3X79WPL2AfY8frrAdPRMAyjghkNvCQia0VkLfAi7pPyKk11NPqTgfdU9W1VfR+YKyJ1g61UVSAtM4fWOevJIwQadwnotdvEtuG1Ia+Rm5fLld9eyZa0LX7ThURG0vqF56l32mlsf+pptj3zrM3eZxhGtUNVF6vq4UBPoKeq9gZODLJapVIdjf50oI7PcR3g+yDpUqXYnJxBB9nC3rotILxO6RkqmE4NOjF+8HhSslK48tsr2bF3h990Eh5OyyefoMH555P06qvs+XJagDU1DMOoGFR1j8+c+7cEVZkyUB2NfpSqpuYfePvW0we2JGfQXHaSG1t5n+uVRnxcPK+c/Apb07dy1XdXkZzpf2kECQ2l+QNjafnkE9Q7dWiAtTQMw6gUqvz849XR6KeJyBH5ByLSh2oQPBEINu5OpwVJhDZoHVQ9ejXtxXMnPMfa5LVc+/21fhfoAfe9f/0zzkBCQ8neupUtDz5EXlZWgLU1DMOoMKr8WGV1NPo3Af8TkR9FZA7wEeB/LthaxqptqZye9xRRZz4VbFU4puUxjBs4jmVJyxgzYwwZORklpk//9Vd2ffABiaNHk5eeHiAtDcMwykf+ei9+thSg6LrjVYxqZ/RVdR5wCHANLnryUFVdUJa8IjJURFaIyCoRucuPvJ2ITBeRJSIyS0Ra+8hGishKbxvpc76PiPzhlfl8MFf/W7ktldZNGhEa3ShYKhTgxLYn8shxjzB/y3xu/eFWsnOLrsyXT/0zz6TFo4+SNvdX1l/xL3L3VPkprA3DqIWoaqyq1vOzxapqlV/PptoZfRG5DohW1aWquhSIEZFry5AvFHgJOBWIBy4UkcIT+owD3lHVnsCDwKNe3kbAf4CjgH7Af0SkoZfnFeBKoIu3BW2AOn3zSu5gIuxMCJYKRRjWcRj3HXMfsxNnc9ePd5GTl1Ns2gZnjaDVs8+wd+lS1o0cRU5SUgA1NQzDqPlUO6MPXKmqu/MPVHUXzuiWRj9glaomqGoWMAkYXihNPJA/kfxMH/kpwHequtO73nfAUBFpAdRT1bnqvjt7B7fMYsBJzcyh9jEN4wAAIABJREFUQepKBu2aDBn+g+eCxXldz+O2vrfx7bpvuWfOPeTm5Rabtt6QIbR55RVCIiORsCrfaDYMowpRBm/u8SLyu4jkiMi5hWS5IrLI26YGTuvAUh3fqqEiIp6Rze/BR5QhXyvAd/L3RFzP3ZfFwNnAc8BZQKyIxBWTt5W3Jf5/e/cdHlW1NXD4tzIzyaSQHiCQYBKagHSkK4iKKF0Q4RMURVQsiPVeGxcRy/VawQpSRYqgKCpdiA0UovQahAAJJSE9pE5mf3/MIQQMISFlErLf55knM6euOYGss/fZpYjl/yAiDwAPALi6liTc0jkYn0FtMe6FagWX+/HL6p4W92Cz23jvr/dwERemdJuCycVU5LZe3bvh2a0rIoI9JwdbwmlcQ5zXI0HTtKqvUG3uzTj+Fm8RkeVKqT2FNjsKjAaeLuIQWVV9CN3yUB1L+quAxSJyo4jcCCwEVpbTsZ8GeojIVqAHEAdcvFhaCkqp6UqpDkqpDuYKKMFGn0on4OxUzh5Vc1z7MS3HML7teL4/9D0TN04stsR/tmnEyZcnEzN8ONn791dWmJqmVU+XrM1VSsUopXYAdmcEWBVUx6T/LxxV8A8Zr52cP1jPxcQBoYU+hxjLCiiljiulbjdGVnrBWJZSzL5xxvuLHrOyRMdnEGRKR1l9wWRxRgglMrbVWB5p8wjL/17OpE2TsKvi/+8FjLkPMZk4cvc9ZG7dWklRappWRZkLTWUbZdSgnnWxGtmSshrH/F1EnPKYtjJUu6SvlLIDfwAxOO7segF7S7DrFqCxiISLiCswHDjvuY2IBIrI2WvyHI6pEwFWA71FxM9owNcbWK2UOgGkiUhno9X+3cC3ZfqClyn6VDpBVjviGeSM05fKQ60fYlzrcXxz8Bsmb5pcbOJ3a9iQq774ApOvD0fvvY/09esvuq2maVc829kaU+M1vRyPfZVSqgPwf8B7ItKwHI9dZVSbpC8iTUTkPyKyD5iG49kMSqkblFIfXGp/pZQNR3/+1ThuEr5USu0WkckiMsDYrCewX0QOAHWAV419k4BXcNw4bAEmG8sAHgY+Aw4Cf1N+jxpKJTo+g+/DX4RH/nDG6UttXOtxPNDqAb6K/opJGycVW9XvGlKfsAULcGvcmBMvTcReQdMSa5pWrV2yNrc4Sqk44+chIBJoW57BVRXVqSHfPhzT6PZTSh0EEJEnSnMApdQKYMUFyyYWer8Ux4Q+Re07i3Ml/8LLo4CiJ5GvJPl2xYnUbAb6u8NFGsdVNSLCo20exUVc+GT7J+Tk5/Bq91cxuxT9T9IcEMBVc2aTGxuHi6dnwSQ9ThwWQdO0qqWgNhdHsh+Oo9R+SUYNbqZSKkdEAoFuwJsVFqkTVZuSPo5W9SeADSIyw2jEp//iA6czcsi3KwYd+y/sLPKepUoSER5p8wiPt3ucFYdX8OzPzxY7gI+LpyfWpk0AOP3hR5ycOBFlu3i/f03Tao6S1OaKyLUiEgvcAXwqIruN3ZsBUSKyHUd37TcuaPV/xag2JX2l1DfANyLiiaNF5gSgtoh8DCxTSq1xaoBOdDwlC8FOo7hvID7C2eGU2v0t78fN5MabW94kLz+Pt3q+hZvJrdh9lC2PlCVLsSWcpv677+DiXvmzCmqaVrWUoDZ3C+c3vj67fCPQssIDrAKqU0kfAKXUGaXUAqVUfxy/vK04WvTXWCdTs/ElA1F2qAYN+YoyqvkoXuz0IpGxkYxfP54sW/FzKNWeMIG6/5lIxk8/cXT0vdiSkyspUk3TtOqr2iX9wpRSyUb/9xudHYsznUjNxl/SHR+qaB/9krjz6juZ3HUym45v4pEfHyEzr/iJd/xGjKD+1PfJ3ruXo3ffjcq7+KMBTdM0rRpV72sXdyI1i0CTUTJ293VuMGU0uPFgXE2uvPDrC4xdM5aPbvoIHzefi27vffPNmGfPIi8uDrFU3fEJNE3TqoJqXdLXHE6m5VDHywU8a4NH1Zhhryz6RvTlnZ7vsC9pH6NXjSY+M77Y7T3at8dngKPXZdqaNaStWFHs9pqmaTWVTvpXgLSsPI54tYVnoqF+e2eHUy56NejFxzd9zPGM49y98m6OpR275D5KKVIWf0nck0+R8NFHBd36NE3TNAed9K8AGTk2almvvKrtjsEdmXnLTM7kneHuVXezP6n48fdFhJCPP8J7QH9OT53G8X/9C3tubiVFq2maVvXppH8FSM/O48bc9bBwBNivrHkkrgm8hrl95uIiLty7+l62xW8rdnsXV1fq/fe/BD0+nrTl33F09L3Ys7MrKVpN07SqTSf9K0BGto2I/ENw6CdwufJ+pRG+EXx+6+f4W/15YO0D/BL7S7HbiwiB48ZR/913cG/TBhertZIi1TRNq9quvAxRA6Xn2PBWGdW+5X5x6nnVY06fOYR5h/HY+sdYeuDSIw9633ordZ59BoDsffvI+O23ig5T0zStStNJv5pTSjme6ZMB1is36QMEugcyu89sutTrwsubXub9v96/5NS8Z8W/+y7HHniQpAULdAM/TdNqLJ30q7nM3HyUAg/7lV3SP8vT4sm0XtO4o8kdfLbzM/7987/Jzb90Y736b7+NZ/dunJr8Csef/ZeeqU/TtBpJJ/1qLiPHMeFMnjUIAhs7OZrKYXYx81Lnl3ii/ROsjFnJ2DVjSclOKXYfk5cXoR99ROD4x0j74QcO3zGMvFPF9//XNE270uikX82lZzuS/vYu70H/950cTeUREe675j7+d/3/2HV6F6NWjrpkX34xmQh6+GEazJqJW9MmmAOq/0BGmqZppaGTfjWXnu0Yb76WW80cUblPeB9m9J5Bck4yI1eOZHvC9kvu49m5MyHvvouYzdhOn+bUf9/U3fo0TasRdNKv5jJybASQSse1t8OB1c4Oxyna1WnH/Fvn42nxZMzqMaw7sq7E+2b8+itJs2cTc+dwcg4frsAoNU3TnE8n/WouI9tGPUnEM3En2G3ODsdpwnzCmH/bfJr6N+WJyCeYvmN6iVrp+w4aROj0T7GdOkXMkKF63H5N065oOulXc+k5NoIl0fHBu75zg3Eyf6s/M3vPpG9EX6ZtncazPz9Lli3rkvt5XX894cu+xq1JE+KefIrkL7+shGg1TdMqn0761VxaVh7BkuT4UMOTPoDVbOX17q/zRPsnWB2zmntW3sPJMycvuZ8lOJirPp9H4PjH8O7dGwClx+3XNO0KU6OSvoj0EZH9InJQRP5dxPoGIrJBRLaKyA4Ruc1YfpeIbCv0sotIG2NdpHHMs+tqV+Z3OpmaTagpGWVyBc/Ayjx1lXW2Zf+0XtM4mn6UO7+/k41xGy+9n8VC0MMPY/L1RdlsxIwcxcnXXsOedenaAk3TtOqgxiR9ETEBHwK3As2BESLS/ILNXgS+VEq1BYYDHwEopb5QSrVRSrUBRgGHlVKFZ3656+x6pVSldv4+kZpNjjUIadwbRCrz1FVej9AeLOi7AH+rPw+te4hpW6eRb88v0b7KZsP9mmtInvc5hwYNIvPPPys4Wk3TtIpXY5I+0BE4qJQ6pJTKBRYBAy/YRgHexnsf4HgRxxlh7FslnEjNYmPtYTD8C2eHUiVF+ESwoO8CBjUaxPQd0xm7diwJmQmX3M/FaqXuxJdoMGcO2PI5MnIUp15/XXft0zStWqtJSb8+UHj0llhjWWGTgJEiEgusAB4r4jh3AgsvWDbbqNp/SaTo4raIPCAiUSISZbOVXyv7E6nZ1PV2L7fjXYncze5M7jaZV7u/yq7Tuxj63VA2Hd9Uon09O3ciYvm3+I0Y7piwR9emaJpWjdWkpF8SI4A5SqkQ4DbgcxEpuEYi0gnIVErtKrTPXUqplsB1xmtUUQdWSk1XSnVQSnUwm8tnIB1bvp349ByeOPYYrHmxXI55JRvQcAAL+y7Ez82PB9c+yIfbPixRdb+Lpyd1J04kfMkSXNzcsJ85Q8KHH+pSv6ZVMSVot3W9iPwlIjYRGXrBuntEJNp43VN5UVeumpT044DQQp9DjGWFjQG+BFBKbQKsQOHWccO5oJSvlIozfqYDC3A8RqgUpzNyybcrArOPQG5mZZ22Wmvo25AFfRfQv2F/Ptn+CQ+ufZDTWadLtK+Lu6NGJT0yktPTPuDQgIGkr9+gZ+3TtCqghO22jgKjcfytLryvP/AfoBOOv+H/ERG/io7ZGWpS0t8CNBaRcBFxxZHAl1+wzVHgRgARaYYj6ScYn12AYRR6ni8iZhEJNN5bgH7ALirJseRMQOFqS68RM+yVFw+LB692f5VXur3C9oTtDF0+lD9O/FHi/X369qXBnNmI2Uzsww9zbMz95ERHV2DEmqaVwCXbbSmlYpRSO4AL5+S+BVirlEpSSiUDa4E+lRF0ZasxSV8pZQMeBVYDe3G00t8tIpNFZICx2VPAWBHZjqNEP1qdK8ZdDxxTSh0qdFg3YLWI7AC24ag5mFEJXweA6FMZeJGFi8oHq076pTWo0SAW9F2At5s3Y9eM5X9b/kdOfk6J9vXs3JmIb7+hzvPPk7VrF6feequCo9U0DTCfbRtlvB4otK4k7bYupiz7Vis1apYWpdQKHA30Ci+bWOj9HqDbRfaNBDpfsOwM0L7cAy2h6Ph06liM58q6pH9ZGvs1ZlHfRbzz5zvM2zOPX+N+5bXrXqNFQItL7isWC/53j8K7fz9UpuPxSl5cHOmRkfjdeSdSTm03NE0rYFNKdXB2ENVZjSnpX4miT2XQILAWNB8IAY2dHU615WHx4MXOL/LJTZ+QkZvByB9G8uG2D0tc6jf7+WGp7ygUpHz7LademcLhwYM5s/HSAwJpmlZuStJuqyL2rVZ00q/GouPT8QsOg2Hz4Kouzg6n2utWvxtfD/yaPuF9+GT7JwxZPqRUz/oBAseNI+SDadizczh63xiOjXuY7P37KyhiTdMKKUm7rYtZDfQWET+jAV9vY9kVRyf9aio7L59TaTmEBXg6O5Qrio+bD69f9zqf3vQp+fZ87l9zP8/98hyJWYkl2l9EqHXTTUR8/x1BTz5J5pYtpCzWE/hoWkUrSbstEbnWGIflDuBTEdlt7JsEvILjxmELMNlYdsUR3d2o8nl6eqozZ86U6RhHEzO5/n8b+LLTITrufwse3gTe9copQg0g25bN9B3Tmb17Nh5mD55s/ySDGw/GRUp+r5yfmoqy2zH7+ZG5ZQvJCxcS+MgjuDVsWIGRa9qVSUQylVK6pFMGuqRfTR1PdUwCE+SSDtkp4FbLyRFdeaxmK+Pbjeer/l/R2K8xkzZNYvSq0RxMPljiY5h8fDD7Obr75h49RnrkTxzq15+4p58h59Dhigpd0zStSDrpV1MnUx2t9n1VGpjcwNXLyRFduSJ8I5h9y2wmd53M4dTD3PHdHbz353tk5pVuQCTfIbfT6Md1BNw/hvQff+RQv36cfO21Copa0zTtn3TSr6ZOGEnfKz/FMaWuHhO+QokIgxsPZvmg5fSN6MvMXTPpt6wfy6KXYVcXjvNxcWY/P2o/9RSN1q3Ff/RoXENCAMesflk7dlRU+JqmaYBO+tXWidQsvK1mLNmJ4BHg7HBqDD+rH1O6T2H+bfMJ9gxm4saJDP9+OFtObinVccwBAdR59hn8774bgLTVq4kZdicxI0eSvmEDyl7yGwlN07SS0g35nKA8GvLdPzeKY0mZrO68E2w5cN2T5RSdVlJKKVYcXsF7f73HyTMn6VqvK4+2eZSWQS1LfSz7mTOkLF1K4ty52I6fwLVhQ/zvvhvfIbfrQX40zaAb8pWdTvpOUB5J/8a3I2lU24tPR+nBqZwt25bN4v2L+WznZ6TkpNAztCePtnmUpv5NS30slZdH2qpVJM6ajcrJIeKH7xERbElJmP39KyB6Tas+dNIvO530naCsST/XZqfZxFWM69GQp3tdBRZrOUanXa4zeWeYv2c+c3fPJT0vnT5hfXig1QM09iv9aIlKKfKTkjAHBGDPzCS6R0/cmjbBb/gIavW+GRdX1wr4BppWtemkX3b6mX41dPj0GfLtiqsDTPBqHfhtqrND0gBPiycPtn6QlUNWMrblWH6K/Ynbl9/OY+sfY3vC9lIdS0QwBxhtNZQi8KEHsZ2K5/jTT3Ow5w3Ev/0OeadOVcC30DTtSqaTfjUUHZ8OQNNauY4FerKdKsXHzYfx7cazZsgaHm79MFvjtzJyxUjuW30fG+M2UtraNRdPTwLGjKHh6lWEzpiBe9u2JM6ciS0+AQBbUhL2nJLNE6BpWs2mq/edoKzV+1N/jOadtQc4cJ8V1wW3w11LofHN5RihVp4y8zJZemApc/fMJT4znuYBzbm7+d30vqo3FpPlso5pS0jAFBiIiHD8hRdIX/cjPv374zt0CNarry7nb6BpVYOu3i87nfSdoKxJ/99f7WDd3nii2vwA2xfBM3+Dq0c5RqhVhNz8XL4/9D2zd80mJi2GIPcghjUdxtAmQwl0D7zs42Zu2ULyosWkr1mDysvD2rw5vsPvxG/YsHKMXtOcTyf9stNJ3wnKmvTvmbWZ1DNZfJM9Bq7qBsPmlmN0WkWzKzu/xf3GF/u+4Le437C4WLg1/FaGNx3ONYHXIJc50JItOZm0774ndfly3Nu0oe6LL6DsdpLnz8erV6+CgYA0rbrSSb/sdNJ3grIm/d7v/kRDfzc+bhcLPiHQoHM5RqdVpsOph1m4byHfHvyWTFsmTfyacHvj2+kX0Q8fN5/LPq6y2RCzmazdu4kZMhQA6zXX4N3nFmr16aNvALRqSSf9stNJ3wnKmvRbTlrN7W3r8/LAa8oxKs2Z0nPTWXl4JV9Ff8WexD24urhyc9jNDGk8hA51Olx26R8gNzaW9NWrSVu1muydOwG4av7neHTogMrPR0ym8voamlahdNIvO530naAsST8jx8Y1/1nNtE6p9L+2KYS0L+foNGfbm7iXr6O/5odDP5Cel06DWg0Y0HAAt0XcRmit0DIdOzc2jvS1a/G/6/8QV1fi332PM7/+Sq0+t+Ddpw+uoWU7vqZVJJ30y65GJX0R6QO8D5iAz5RSb1ywvgEwF/A1tvm3UmqFiIQBe4H9xqa/K6UeMvZpD8wB3IEVwOPqEhe1LEn/YHw6N73zM9trT8InuDGMWHBZx9GqvmxbNmuPrOXr6K+JOhUFQKugVvQN78stYbcQ4F72ORdSln1D8sKFZBuT/bg1aYL3bbcR+NCDZT62ppU3nfTLrsYkfRExAQeAm4FYYAswQim1p9A204GtSqmPRaQ5sEIpFWYk/e+VUv+oTxeRzcB44A8cSX+qUmplcbGUJelv2BfPvXO2cMD/KVyb9IJBH13WcbTq5UTGCVYcXsEPh38gOjkak5joXK8zfcL60COkB35WvzIdPy8ujrTVa8iIjMQcGEj9d94GIGHqVKwtWuDZuTMuns77W5uXl0dsbCzZ2dlOi0GrPFarlZCQECyW87u06qRfdjVpJo+OwEGl1CEAEVkEDAT2FNpGAd7Gex/geHEHFJFgwFsp9bvxeR4wCCg26ZfF2YF5LHmpYNWD8tQUwV7BjGk5hjEtx3Ag+QArDq1gxeEVvPTbS7iIC22C2tCrQS9uCL2BBt4NSn18S/36BNx3LwH33Vsww58tOZmkufOwnzkDFgse7dvjdf311OrdG9eQ+uX9FYsVGxtLrVq1CAsLK1P7Bq3qU0qRmJhIbGws4eHhzg7nilOTkn594Fihz7FApwu2mQSsEZHHAE/gpkLrwkVkK5AGvKiU+sU4ZuwFx6zQv4bRpzKo42lC8jL1SHw1VBO/JjRp34TH2z3OnqQ9bDi6gQ3HNvBW1Fu8FfUWDX0a0jO0Jzc0uIGWgS1xkdINvCkuju3Nfn402bSRzL+2kvHLz5z5+Rfi33wTk78friH1yTtxgswtW/Ds0gVzUFBFfNUC2dnZOuHXECJCQEAACQkJzg7lilSTkn5JjADmKKXeFpEuwOcicg1wAmiglEo0nuF/IyItSnNgEXkAeADAtQyTpUTHZ9AmEDiFLunXcCJCi4AWtAhowaNtHyU2PZbIY5FEHotkzu45zNw1k0D3QHqE9KBXg150rNsRq7l0kzOJqyuenTvh2bkTPPMMeceP4+LtqAzLiIzk5MuTAXBr2hTPrl3x7NoFj06dKmRCIJ3waw79u644NSnpxwGFmyaHGMsKGwP0AVBKbRIRKxColIoHcozlf4rI30ATY//CHZ6LOibGftOB6eB4pn85X0ApxcH4DNq1qQP9fwTvyq1i1aq2kFohjGw+kpHNR5Kak8ovcb+w4eiGgq6A7mZ3utbryg2hN9CtfrfLGgXQUq9ewXvfYcOwXtOSMxs3cmbjRpLnzydpzhya/L4JXF3JjIpC5ebi3rq1U9sDlIfExERuvPFGAE6ePInJZCLIqN3YvHlzsTfyUVFRzJs3j6lTi58Yq2vXrmzcuLHcYp4wYQJLlizh2LFjuLjoaVY0h5rUkM+MoyHfjTgS8xbg/5RSuwttsxJYrJSaIyLNgB9xVNcHAklKqXwRiQB+AVoqpZKKaMg3TSm1orhYLrchX9KZXNq9spaX+jVnTHf9rEsrmdz8XDaf3MyGoxuIPBZJfFY8AI18G9E5uDOdgzvToW4HPC1lS8z2zEyy9+3Ho11bAI7eP5Yzv/4KJhPWZs3waN8ez+7d8bque6mPvXfvXpo1a1am+MrLpEmT8PLy4umnny5YZrPZMJurThnKbrcTHh5OcHAwr7/+OjfccEOFnKciv3dRv3PdkK/sasztn1LKBjwKrMbR/e5LpdRuEZksIgOMzZ4CxorIdmAhMNrofnc9sENEtgFLgYeUUknGPg8DnwEHgb+pwEZ8J1KzAGhoPg1/fQ6ZSZfYQ9PA1eRK9/rdeanLS6y7Yx2L+i1iQrsJBLoHsuTAEh5d/yjdFnZj1IpRfLD1A6JORpGXn1fq87h4eBQkfID6771L6IwZBIy9HxerleSFC0n89NOC9aenzyB1+XJyY2JKPfNgVTB69GgeeughOnXqxLPPPsvmzZvp0qULbdu2pWvXruzf7+jhGxkZSb9+/QDHDcN9991Hz549iYiIOK/07+XlVbB9z549GTp0KFdffTV33XVXwfVZsWIFV199Ne3bt2f8+PEFx71QZGQkLVq0YNy4cSxcuLBg+alTpxg8eDCtW7emdevWBTUL8+bNo1WrVrRu3ZpRo0YVfL+lS5cWGd91113HgAEDaN68OQCDBg2iffv2tGjRgunTpxfss2rVKtq1a0fr1q258cYbsdvtNG7cuOB5vd1up1GjRuX2/F5E+ojIfhE5KCL/LmK9m4gsNtb/YfTMQkTCRCRLRLYZr0/KJaAqqOrcmlYCowS+4oJlEwu93wN0K2K/r4CvLnLMKKBShsY7merorhSWtRtWPQ6hncDDvzJOrV0hCrcDGNNyDDn5OWyL38YfJ/7gjxN/MGPnDD7d8SnuZnfaBLWhfZ32tK/TnpZBLXEzuZXqXCYvL7yuO1eyt+fkkH/6NAAqL4/EmTOxp6Y6tvX1xdqqJb5DhuJ9S+9ij/vyd7vZczztMr79xTWv581/+peqmQ7g6FWwceNGTCYTaWlp/PLLL5jNZtatW8fzzz/PV1/988/Gvn372LBhA+np6TRt2pRx48b9o2va1q1b2b17N/Xq1aNbt2789ttvdOjQgQcffJCff/6Z8PBwRowYcdG4Fi5cyIgRIxg4cCDPP/88eXl5WCwWxo8fT48ePVi2bBn5+flkZGSwe/dupkyZwsaNGwkMDCQp6dKFib/++otdu3YVtK6fNWsW/v7+ZGVlce211zJkyBDsdjtjx44tiDcpKQkXFxdGjhzJF198wYQJE1i3bh2tW7cueFRSFka37A8p1C1bRJYX7paN4xFuslKqkYgMB/4L3Gms+1sp1abMgVRxNSrpV3fHjaTvb8p0LHAvW99sTXMzudEpuBOdgh0dWdJy04g6GcXvJ37nz1N/8uG2D1EoLC4WWga2LLgJaFO7TakfB7i4ueFS39EORSwWmmz8jZyDf5O1fRtZ27eTvWMHtpMnAMiLj+foPaNxb9UK9zatoVYtVKNGiOXypiKuKHfccQcmYxjj1NRU7rnnHqKjoxER8vKKri3p27cvbm5uuLm5Ubt2bU6dOkXIBXMhdOzYsWBZmzZtiImJwcvLi4iIiIJEO2LEiPNK1Wfl5uayYsUK3nnnHWrVqkWnTp1YvXo1/fr1Y/369cybNw8Ak8mEj48P8+bN44477iAw0NHGw9//0gWJjh07ntedburUqSxbtgyAY8eOER0dTUJCAtdff33BdmePe9999zFw4EAmTJjArFmzuPfeey95vhIqSbfsgTh6aYGj1vYDqWGtBnXSr0ZOpmZhchG8bCmOBdbLn5BF04ri7epNrwa96NWgFwCpOalsjd/Kn6f+5K9TfzFr1yxm7JyBi7jQ0LchrQJb0SqoFS0DWxLhE4HJpeTj+IvJhLVpE6xNmxRMA3y2GltlZuIaHk7Gr7+S+u23AOx/8SVCpr7Pf/pfjy0pCVt8PG4NGzr1RsCzUAPFl156iRtuuIFly5YRExNDz549i9zHze1cjYnJZMJms13WNhezevVqUlJSaNmyJQCZmZm4u7tf9FHAxZjNZuzGmA12u53c3NyCdYW/d2RkJOvWrWPTpk14eHjQs2fPYgdRCg0NpU6dOqxfv57NmzfzxRdflCquYpSkW3bBNkopm4ikAmeHtiyqW/YVRyf9auREajZ1arnhkhgNvleBufy7RWlaYT5uPvQM7UnP0J4AZOZlsj1hO3/F/8XOhJ2sPbKWr6IdVdieFk+uCbiGlkEtaRXYipZBLUvdQ+Bsocs1LIzQjz5EKUVe3HH+XrECv8REXCMiAEhft46TE/+DuLnhdnVT3Ftcg7VlS7x73+y0ngKpqanUN2oy5syZU+7Hb9q0KYcOHSImJoawsDAWL15c5HYLFy7ks88+K6j+P3PmDOHh4WRmZnLjjTf63By5AAAXW0lEQVTy8ccfM2HChILq/V69ejF48GCefPJJAgICSEpKwt/fn7CwMP7880+GDRvG8uXLL1pzkZqaip+fHx4eHuzbt4/ff/8dgM6dO/Pwww9z+PDhgur9s6X9+++/n5EjRzJq1KiCmpISMotIVKHP042eUWVVZLdspVT5PkeqAnTSr0ZOpGQT7OsOCfuhdtVoyazVLB4WD7rU60KXel0AR8n8SNoRdpzewY6EHew8vZM5u+ZgU46SaT3PerQMaknLwJY082/G1QFX4+3qXdwpziMijtH/rutOnUItub169KTe//5H9u7dZO/aReo335C8YAG1bugJQMpXX5O1cwfWZs2xNrsat0aNcPHwKL8LUYRnn32We+65hylTptC3b99yP767uzsfffQRffr0wdPTk2uvvfYf22RmZrJq1So++eRcOzRPT0+6d+/Od999x/vvv88DDzzAzJkzMZlMfPzxx3Tp0oUXXniBHj16YDKZaNu2LXPmzGHs2LEMHDiQ1q1bF5yzKH369OGTTz6hWbNmNG3alM6dHVN9BwUFMX36dG6//Xbsdju1a9dm7dq1AAwYMIB77733cqr2bUqpDhdZV5Ju2We3iTV6dPkAiUaD7aK6ZUdxhakxXfaqksvtsvfZL4cwuwij2/pATjr4XVUB0Wla2WTbstmbtJcdCeduBE6cOVGwPsQrhOYBzWkW0Izm/s1p6t/0kpMHXarLnsrPJ/fIUdwiHM+PE6ZOJenz+djTHcNWI4Jb06aEL/saESFr125c3K24XnUVUoW62l1KRkYGXl5eKKV45JFHaNy4MU888YSzwyq1qKgonnjiCX755eI16KXtslfCbtmP4Ohu/ZDRkO92pdQwEQniIt2yy/ZNq57q869d4/7rIs590K32tSrKarbStnZb2tY+130vKTuJfYn72JO0hz2JjteaI2sK1ge6B9LUr6ljiGH/JjT1a0qYTxgWl5I9rxeTqSDhAwSNH0/gY4+RFxtL9t695ERHo7JzCh4fnHrlFbK2b0csFlwbNsStSWM8O3bEd+jQcroKFWPGjBnMnTuX3Nxc2rZty4MPVr/ZEN944w0+/vjj8nyWDxQ8oz/bLdsEzDrbLRuIUkotB2biGGn1IJAEDDd2vx6YLCJ5gJ3zu2VfUXRJ3wnKMsseJ3fC3u+h41jwLP2IappWVaTmpLIvaR/7k/ZzIPkAB5IPcDDlIHl2x7Nji4uFhr4NaeLXBK8cLzo37kyEbwQhXiGlajBYlOz9+wtuBnIORJNz4ADWltcQ+sEHABzq3x8sFtzCwnEND8c1Ihz3Fi1wDQsr69fWSkgPzlMxdNJ3gjIl/ahZ8P0T8MQe8NHD8GpXljx7HjGpMexPNm4Ekg4QnRxdMIogOG4GwnzCiPCJOPfyjSDMOwxX0+U3blV5eYjFgsrP59Trb5B75Ai5hw+TFxcHSuF3113UfelFVF4eseMfxzUiHLfwcCyhDXBtEIq5Tp2CyYq0stNJv2Lo6v3qJjUOxAS16jo7Ek0rdxYXC439GtPYr/F5y//c+SeWuhYOpR7iUMohDqUecjwiiFmDwlFwcREXQrxCiPA9dzPQ0Lch4T7hJRpT4GzXPzGZqPviCwXL7dnZ5B45gou7O+CYcjjv+HHO/PorqlCL9trPPE3AmDHYkpJImjsP16uuwjUsDNcGoZgCAvQkMlqVoJN+dZN23JHwy1i9qWnViYfZg2ZBzWgV1Oq85dm2bI6kHeFQ6iH+TvmbQ6mHOJx6mF/jfsVmP9e3vY5HHcK8wwj1DqVBrQY0qNWAUO9QQmuF4m52L/bcLlYr1qZNCz5batcm4ttvUDYbeSdPknf0KLlHj+He1tGGITcmhsTPPoP8/HPH8PCg/rvv4NWjB7mxsZz59TcsISFY6tXDUi8YF2vpZj/UtMulk351kxYH3vUuvZ2m1QBWs5Wm/k1p6t/0vOV59jxi02PPqxk4mn6UH4/8SHJO8nnb1navfe5mwLsBobUc70NrheLl6nXRc4vZjGtICK4hIXh2Pbfco107rt62lby4OHJiYsg7eozcY8ewhDYAIHNLFCcnTTrvWCZ/fxrMmon16qvJ2rWbrL/+wlK/nnFTUA8Xb29dU6CVC/1M3wnK9Ez/oy4Q2BiGzSvfoDStCivPWfbSctM4ln6MY2nHOJp+lKNpRzmW7nh/Ouv0edv6W/0J8Qoh0DWQDd9sgDRIjkmGFPC3+CN2ueTUuuAYtc7V1ZWuXbui8vOxxceTFxdH3vHjjlfccZ7e+hd/nz7N9/feR8K77563v4unJw1XrcQcFETGr7+Rs28vlnr1MAcHY6lXH3NQ4BXXnkA/068YuqRf3Tz0G9iynB2FplVb3q7eBZMOXehM3hli02PPuxmIzYjlQNoBTJ1M2JQNX3wBUCiCPIIY++NYQmqFUNezLsGewQWvup518bA4BgSKjIzEy8uLrl27IiYTluBgLMHBBedNSUnh55Yt8fLyIvWmG2k8dEjBzUDeiRPkHT+OyRjNLuPnn0ie9/l5cYu7O02jtiAmk2PmwqPHHLUEdetgrlMHS926uHh6VrkpgLXKp0v6TlCmkr6m1UDlWdK/XDa7jYTMBKZ8MIV8z3wCGwaycuNK8jzyEF8BT7BjP28fU66JBn4N2LdlHypV4W5zZ/SQ0fRo14Pa7rUJ9AjE3ezOrFmziIqKok6dOlgsFp5//nkADh48yEMPPURCQgImk4klS5bQsGFD3n7lFX5cvJjaCD2aN6d/rxsYumgRb731FsFffkna8u/OiyPFbOaF2kHk5+ezYMAAVn/xBady80i027n1rv+j2+DBWJs3Z968ebz11luICK1ateKjjz6iVatWHDhwAIvFQlpaGq1bty74XJF0Sb9i6Fu+6iTpMPzyFnR5VA/Dq9Vss4sY5rbFIMf4FbmZ8MUd/1zf5v+g7V1wJhG+vPv8dff+cMlTml3MBHsFE5QdhKfJk2VTlrHq21UEBQWxePFiVq1exWtTX6PTzZ2YOmcqp3NPcyTpCEm2JJIaJpFlySLXJZfZ8bOZvWp2wXFrudYiKz2LRj0bYfIyEfl9JFftvYraHrV57snneGTcI4wcNJL8vHzsdjsrV67kq5UrWbd5Mx4eHufGtF+0CID6b76J64QJDOjWjXWLF7N+yRK+WrKEpT//hL+/P7FPPUU3Ty/s2Ylgt8Pn8zm5azdnXnyBKVOmsKJde1zOZKAQ0qe8ysTwCH5647/c9NKLLFq0iFG33orJZoMqNuOhVjI66Vcnp6Nh63xod4+zI9G0Gi0nJ4ddu3Zx8803A5Cfn09wcDDBXsE092nO/BfmM2jQIP416F94eXkx6ZdJeHl5Me7xcZzIOEF8VjwJmQkkZCUQczqGpduW4trIlf2Z+6EDvLH5DceJhsA7Ge/wzvx38Lf6E+QeROKRRELvD2XW/lkEuQcR4B5AoC2Q/Fr55NhzABBXV04ohWfHjiTu2YOcPl0w2U2dN95wDIH700/4WsykHo7h++mfsn79eu644w68/QPIOXwIW0ICZzZvplN8PDuWfwsvvcjs2bOZmZfH/rbtcPH0xBQQgNnfn1p9biFg9GgAkhYswOzri8k/AHOAP6aAAEw+PkjpJtbRKohO+tVJmjF3hG69r9V0xZXMXT2KX+8ZUKKSfXGUUrRo0YJNmzb9Y90PP/zAzz//zHfffcerr77Kzp07z53a4kkjv0Y08mtUsGzatGmc+OQEvy3+DYC09DTue+Q+RowdwZB7hvD2J2+fu0nITOCE+QRHXI6wb+c+7KrQ44Sh8NDeh3CPdsfX4ovHGA+e2PAECSSQ0jiFxfsWE+AewO8bfif2TCyRf/yMr7sv4eHh2Bo0gO3bAaj91JPnf1e7nXvatKFuZCT5Nhv1XngR26lT2JISyU9MwpaUCPnGFLyZmZya/Mo/rknA/WOo/fTTl3/BtXKjk351khYH4gJeemAeTXMmNzc3EhIS2LRpE126dCEvL48DBw7QrFkzjh07xg033ED37t1ZtGgRGRkZ1KpVi7S0omdpXbhwIatWraJLF8fMhYcPH+amm27irclvUTezLqY9JsYNGkdOTg75+fn8nPczkydP5pc1v5DjksOhU4fIc83jfx/+j9rhtWnUrBEbt29E2RSHUg9xnONk189myh9THCcUoCdcv+R6zJixPm5lfNR4POt7sv3Ydkw/mQj2CcY135X6/vXxs/rR+77+3PXgXTw34Tl8Bw+66HURd3cab9pIflIStsRE8hMTsSUmYW2uH0dWFTrpVydpxx0J36R/bZrmTC4uLixdupTx48eTmpqKzWZjwoQJNGnShJEjR5KamopSivHjx+Pr60v//v0ZOnQo3377LdOmTeO6664DICYmhiNHjhRMRwsQHh6Oj48Pf/zxB59//jkPPvggEydOxGKxsGTJEvr06cO2bdvo3LEzrq6u3Hbbbbz22msE3h3IsGHDiDJF0bdvX7bP3863L3/LnDlz2By1mZf/+zKJ2YkcPnWYl15/iRxTDoERgRxPPk7txrXJcckhsEUgX+/9GnG/YEwAP/B/3p9P+ZT5C+bj5+aHn9UPXzff8356u3rj6+aLj4cPvn6++LRogLer9yUHQNIqj2697wSX3Xr/6wch9Rjcu6L8g9K0KqwqtN6vSWx2G6k5qaTkpJCUncTKyJVs2raJwf83mOTsZJJzkknJTiE5J5nk7GRSclLIKqYrsauLKz5uPvi4+TCj9wwC3S89WZhuvV8xalSRUUT6AO/jmHbxM6XUGxesbwDMBXyNbf6tlFohIjcDbwCuQC7wjFJqvbFPJBAMnP0X31spFU9FuP3TCjmspmlaYWYXMwHuAQS4B/DeY++xcuVKVqxYQZMmTS66T7Ytm9ScVFJzUx0/jVdKTgqpuamk5aSRkpOCh9mjEr+JdqEak/RFxAR8CNwMxAJbRGS5UmpPoc1eBL5USn0sIs2BFUAYcBror5Q6LiLX4JivufAUd3cppaIq43tomqZVpmnTppVoO6vZitVspY5nnQqOSCuLK2vcxuJ1BA4qpQ4ppXKBRcDAC7ZRgLfx3gc4DqCU2qqUOm4s3w24i4hbJcSsaZqmaeWmJiX9+sCxQp9jOb+0DjAJGCkisThK+Y8VcZwhwF9KqZxCy2aLyDYReUkuMiuGiDwgIlEiEmWz2YraRNO0Yuj2RzWH/l1XnJqU9EtiBDBHKRUC3AZ8LiIF10hEWgD/BR4stM9dSqmWwHXGa1RRB1ZKTVdKdVBKddBjX2ta6VitVhITE3UyqAGUUiQmJmLV0w1XiJqUfeKA0EKfQ4xlhY0B+gAopTaJiBUIBOJFJARYBtytlPr77A5KqTjjZ7qILMDxGEFPgadp5SgkJITY2FgSEhKcHYpWCaxWKyEhIc4O44pUk5L+FqCxiITjSPbDgf+7YJujwI3AHBFpBliBBBHxBX7A0Zr/t7Mbi4gZ8FVKnRYRC9APWFfxX0XTahaLxUJ4eLizw9C0aq/GVO8rpWzAozha3u/F0Up/t4hMFpEBxmZPAWNFZDuwEBitHPWJjwKNgInGs/ttIlIbcANWi8gOYBuOm4kZlfvNNE3TNK1k9OA8TqCn1tU0TSs9PThP2dWYkr6maZqm1XS6pO8EImLn3Ah+pWUGqmKfPx1X6VTVuKDqxqbjKp0rMS53pZQurJaBTvrVjIhEKaU6ODuOC+m4SqeqxgVVNzYdV+nouLSi6DsmTdM0TashdNLXNE3TtBpCJ/3qZ7qzA7gIHVfpVNW4oOrGpuMqHR2X9g/6mb6maZqm1RC6pK9pmqZpNYRO+tWEiPQRkf0iclBE/l0F4okRkZ3G6IRRxjJ/EVkrItHGT79KiGOWiMSLyK5Cy4qMQxymGtdwh4i0q+S4JolIXKFRHW8rtO45I679InJLBcYVKiIbRGSPiOwWkceN5U69ZsXE5dRrJiJWEdksItuNuF42loeLyB/G+ReLiKux3M34fNBYH1bJcc0RkcOFrlcbY3ml/ds3zmcSka0i8r3x2anXSytEKaVfVfwFmIC/gQjAFdgONHdyTDFA4AXL3sQxPwHAv4H/VkIc1wPtgF2XigPHzIkrAQE6A39UclyTgKeL2La58Tt1A8KN37WpguIKBtoZ72sBB4zzO/WaFROXU6+Z8b29jPcW4A/jOnwJDDeWfwKMM94/DHxivB8OLK6g63WxuOYAQ4vYvtL+7RvnexJYAHxvfHbq9dKvcy9d0q8eOgIHlVKHlFK5wCJgoJNjKspAYK7xfi4wqKJPqJT6GUgqYRwDgXnK4XfAV0SCKzGuixkILFJK5SilDgMHcfzOKyKuE0qpv4z36TjmoaiPk69ZMXFdTKVcM+N7ZxgfLcZLAb2ApcbyC6/X2eu4FLhRRKQS47qYSvu3L44ZSfsCnxmfBSdfL+0cnfSrh/rAsUKfYyn+D2JlUMAaEflTRB4wltVRSp0w3p8E6jgntIvGURWu46NG9eqsQo8/nBKXUZXaFkcpscpcswviAidfM6OqehsQD6zFUauQohyTeF147oK4jPWpQEBlxKWUOnu9XjWu17si4nZhXEXEXN7eA54F7MbnAKrA9dIcdNLXLld3pVQ74FbgERG5vvBKpZSi+JJHpagqcRg+BhoCbYATwNvOCkREvICvgAlKqbTC65x5zYqIy+nXTCmVr5RqA4TgqE24urJjKMqFcYnINcBzOOK7FvAH/lWZMYlIPyBeKfVnZZ5XKzmd9KuHOCC00OcQY5nTKKXijJ/xwDIcfwxPna0yNH7GOym8i8Xh1OuolDpl/KG245iC+Wx1dKXGJSIWHIn1C6XU18Zip1+zouKqKtfMiCUF2AB0wVE9bi7i3AVxGet9gMRKiquP8ZhEKaVygNlU/vXqBgwQkRgcjyF7Ae9Tha5XTaeTfvWwBWhstIB1xdHgZbmzghERTxGpdfY90BvYZcR0j7HZPcC3zonwonEsB+42WjJ3BlILVWlXuAueoQ7Gcc3OxjXcaMkcDjQGNldQDALMBPYqpd4ptMqp1+xicTn7molIkIj4Gu/dgZtxtDfYAAw1Nrvwep29jkOB9UbNSWXEta/QjZvgeG5e+HpV+O9RKfWcUipEKRWG4+/UeqXUXTj5emmFOLsloX6V7IWj9e0BHM8TX3ByLBE4Wk5vB3afjQfHs7gfgWhgHeBfCbEsxFHtm4fjWeGYi8WBo+Xyh8Y13Al0qOS4PjfOuwPHH7vgQtu/YMS1H7i1AuPqjqPqfgewzXjd5uxrVkxcTr1mQCtgq3H+XcDEQv8HNuNoQLgEcDOWW43PB431EZUc13rjeu0C5nOuhX+l/dsvFGNPzrXed+r10q9zLz0in6ZpmqbVELp6X9M0TdNqCJ30NU3TNK2G0Elf0zRN02oInfQ1TdM0rYbQSV/TNE3Tagid9DXtCiQi+YVmWtsm5Tgzo4iESaHZAzVNqz7Ml95E07RqKEs5hmjVNE0roEv6mlaDiEiMiLwpIjuN+dgbGcvDRGS9MVHLjyLSwFheR0SWiWPe9u0i0tU4lElEZohjLvc1xqhwmqZVcTrpa9qVyf2C6v07C61LVUq1BD7AMSMawDRgrlKqFfAFMNVYPhX4SSnVGmiHYwRGcAx7+6FSqgWQAgyp4O+jaVo50CPyadoVSEQylFJeRSyPAXoppQ4ZE9ycVEoFiMhpHEPc5hnLTyilAkUkAQhRjglczh4jDMdUro2Nz/8CLEqpKRX/zTRNKwtd0te0mkdd5H1p5BR6n49uH6Rp1YJO+ppW89xZ6Ocm4/1GHLOiAdwF/GK8/xEYByAiJhHxqawgNU0rf/ruXNOuTO4isq3Q51VKqbPd9vxEZAeO0voIY9ljwGwReQZIAO41lj8OTBeRMThK9ONwzB6oaVo1pJ/pa1oNYjzT76CUOu3sWDRNq3y6el/TNE3Taghd0tc0TdO0GkKX9DVN0zSthtBJX9M0TdNqCJ30NU3TNK2G0Elf0zRN02oInfQ1TdM0rYbQSV/TNE3Taoj/B1Nzr8SIgaPIAAAAAElFTkSuQmCC", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "#绘制每个基分类器的训练曲线\n", "for i in range(number_of_weak_classifiers):\n", " plot_loss_accuracy(history[i],\n", " f'{i+1} of {number_of_weak_classifiers} weak_classifiers with {num_of_res_blocks} residual_blocks with {num_hidden_layers_of_res_block} hidden_layers with {num_neurons_of_hidden_layer} neurons',\n", " f'{i+1} of {number_of_weak_classifiers} weak_classifiers with {num_of_res_blocks} residual_blocks with {num_hidden_layers_of_res_block} hidden_layers with {num_neurons_of_hidden_layer} neurons')" ] }, { "cell_type": "markdown", "metadata": { "id": "jFNwxC_SPDvY" }, "source": [ "#### **输出模型结构信息**" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "AK9zeZS7_oQp", "outputId": "ff6ae3d5-4d17-42c1-a8f7-6362305f9af8" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Model: \"model_9\"\n", "__________________________________________________________________________________________________\n", " Layer (type) Output Shape Param # Connected to \n", "==================================================================================================\n", " input_1 (InputLayer) [(None, 2)] 0 [] \n", " \n", " 0th-clf_0th-block_0th-hidden ( (None, 2) 6 ['input_1[0][0]'] \n", " Dense) \n", " \n", " 1th-clf_0th-block_0th-hidden ( (None, 2) 6 ['input_1[0][0]'] \n", " Dense) \n", " \n", " 2th-clf_0th-block_0th-hidden ( (None, 2) 6 ['input_1[0][0]'] \n", " Dense) \n", " \n", " 3th-clf_0th-block_0th-hidden ( (None, 2) 6 ['input_1[0][0]'] \n", " Dense) \n", " \n", " 4th-clf_0th-block_0th-hidden ( (None, 2) 6 ['input_1[0][0]'] \n", " Dense) \n", " \n", " 0th-clf_0th-resBlock_linear (D (None, 2) 6 ['0th-clf_0th-block_0th-hidden[0]\n", " ense) [0]'] \n", " \n", " 1th-clf_0th-resBlock_linear (D (None, 2) 6 ['1th-clf_0th-block_0th-hidden[0]\n", " ense) [0]'] \n", " \n", " 2th-clf_0th-resBlock_linear (D (None, 2) 6 ['2th-clf_0th-block_0th-hidden[0]\n", " ense) [0]'] \n", " \n", " 3th-clf_0th-resBlock_linear (D (None, 2) 6 ['3th-clf_0th-block_0th-hidden[0]\n", " ense) [0]'] \n", " \n", " 4th-clf_0th-resBlock_linear (D (None, 2) 6 ['4th-clf_0th-block_0th-hidden[0]\n", " ense) [0]'] \n", " \n", " 0th-clf_0th-resBlock_Add (Add) (None, 2) 0 ['0th-clf_0th-resBlock_linear[0][\n", " 0]', \n", " 'input_1[0][0]'] \n", " \n", " 1th-clf_0th-resBlock_Add (Add) (None, 2) 0 ['1th-clf_0th-resBlock_linear[0][\n", " 0]', \n", " 'input_1[0][0]'] \n", " \n", " 2th-clf_0th-resBlock_Add (Add) (None, 2) 0 ['2th-clf_0th-resBlock_linear[0][\n", " 0]', \n", " 'input_1[0][0]'] \n", " \n", " 3th-clf_0th-resBlock_Add (Add) (None, 2) 0 ['3th-clf_0th-resBlock_linear[0][\n", " 0]', \n", " 'input_1[0][0]'] \n", " \n", " 4th-clf_0th-resBlock_Add (Add) (None, 2) 0 ['4th-clf_0th-resBlock_linear[0][\n", " 0]', \n", " 'input_1[0][0]'] \n", " \n", " classifiers_Add (Add) (None, 2) 0 ['0th-clf_0th-resBlock_Add[0][0]'\n", " , '1th-clf_0th-resBlock_Add[0][0]\n", " ', \n", " '2th-clf_0th-resBlock_Add[0][0]'\n", " , '3th-clf_0th-resBlock_Add[0][0]\n", " ', \n", " '4th-clf_0th-resBlock_Add[0][0]'\n", " ] \n", " \n", " activation (Dense) (None, 2) 6 ['classifiers_Add[0][0]'] \n", " \n", "==================================================================================================\n", "Total params: 66\n", "Trainable params: 6\n", "Non-trainable params: 60\n", "__________________________________________________________________________________________________\n" ] } ], "source": [ "#打印模型结构和参数信息\n", "boosting_model.summary()" ] }, { "cell_type": "markdown", "metadata": { "id": "mG1YQbHPPFUW" }, "source": [ "#### **对原始特征空间剖分的可视化**" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000 }, "id": "_Nx-bqNC_tNZ", "outputId": "dee69ae1-df6a-4b36-a500-2460d3765c0b" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "79/79 [==============================] - 0s 2ms/step\n", "79/79 [==============================] - 0s 1ms/step\n", "79/79 [==============================] - 0s 1ms/step\n", "79/79 [==============================] - 0s 2ms/step\n", "79/79 [==============================] - 0s 2ms/step\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "#对原始特征空间剖分的可视化\n", "\n", "#可视化原始特征空间\n", "fig, ax1= plt.subplots(1,1, figsize=(7, 4),subplot_kw = {'aspect':'equal'})\n", "mp = ax1.scatter(*X.T,c = y,cmap = cm_bright,edgecolors='white',s = 20,linewidths = 0.5)\n", "#plt.colorbar()\n", "plt.colorbar(mp,ax = [ax1])\n", "plt.title(f'Raw data')\n", "plt.savefig(f'Raw data.png')\n", "plt.savefig(f'Raw data.pdf')\n", "\n", "#可视化每叠加一个基分类器后的EGB模型对原始特征空间的剖分\n", "for i in range(number_of_weak_classifiers):\n", " prob = boosting_models[i].predict(p)[:,1]\n", " fig, ax1= plt.subplots(1,1, figsize=(7, 4),subplot_kw = {'aspect':'equal'})\n", " ax1.scatter(*p.T,c = prob,cmap = cm_bright)\n", " mp = ax1.scatter(*X.T,c = y,cmap = cm_bright,edgecolors='white',s = 20,linewidths = 0.5)\n", " plt.colorbar(mp,ax = [ax1]);\n", " plt.title(f'Outputs of {i+1} classifiers')\n", " plt.savefig(f'Space division of {i+1} classifiers.png')\n", " plt.savefig(f'Space division of {i+1} classifiers.pdf')\n", " \n", "#生成动图\n", "def create_gif(image_list, gif_name, duration=1):\n", " frames = []\n", " for image_name in image_list:\n", " frames.append(imageio.imread(image_name))\n", " imageio.mimsave(gif_name, frames, 'GIF', duration=duration)\n", " return\n", "\n", "def main():\n", " image_list = ['Raw data.png']\n", " for i in range(number_of_weak_classifiers):\n", " image_list.append(f'Space division of {i+1} classifiers.png')\n", " gif_name = '空间剖分动图.gif'\n", " duration = 0.8\n", " create_gif(image_list, gif_name, duration)\n", "\n", "main()" ] }, { "cell_type": "markdown", "metadata": { "id": "Y7nTwYMSPHPo" }, "source": [ "#### **更换背景显示原始样本**" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 409 }, "id": "M0tNx7bR_vJ8", "outputId": "7eb1fede-c697-426f-af39-7915b9ac2e5f" }, "outputs": [ { "data": { "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "#更换背景显示原始样本\n", "\n", "mpl.style.use('ggplot')\n", "\n", "fig = plt.figure(figsize = (9,3))\n", "top = cm.get_cmap('Oranges_r', 512)\n", "bottom = cm.get_cmap('Blues', 512)\n", "newcolors = np.vstack((top(np.linspace(0.55, 1, 512)),\n", " bottom(np.linspace(0, 0.75, 512))))\n", "cm_bright = ListedColormap(newcolors, name='OrangeBlue')\n", "\n", "fig = plt.figure(figsize = (8,6))\n", "m3 = plt.scatter(*X.T,c = y,cmap = cm_bright,edgecolors='white',s = 20,linewidths = 0.5)\n", "plt.title(f'Raw data ({n_samples} points)')\n", "plt.axis('equal')\n", "plt.colorbar()\n", "plt.savefig(f'Raw data ({n_samples} points)')\n", "plt.savefig(f'Raw data ({n_samples} points).pdf')\n", "plt.axis('equal')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": { "id": "SrtYCTwgPIjP" }, "source": [ "#### **特征变换过程可视化**" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000 }, "id": "3CR2-N_8_xka", "outputId": "241f15f6-c35a-4079-d77f-2e1d3940d060" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[, , , , ]\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "#对样本点特征变换的可视化\n", "\n", "clf_add_layers = []\n", "clf_add_layers.append(boosting_models[0].get_layer('0th-clf_0th-resBlock_Add'))\n", "for i in range(len(boosting_models)-1):\n", " add_layer = boosting_models[i+1].get_layer('classifiers_Add')\n", " clf_add_layers.append(add_layer)\n", "\n", "inp = boosting_model.input\n", "outputs = [layer.output for layer in clf_add_layers]\n", "print(outputs)\n", "functors = [K.function([inp], [out]) for out in outputs]\n", "boosting_model_outs = [func([X]) for func in functors]\n", "\n", "#可视化每叠加一个基分类器后的EGB模型对样本点特征变换后的状态\n", "for idx in range(len(boosting_model_outs)):\n", " fig = plt.figure(figsize = (8,6))\n", " plt.scatter(boosting_model_outs[idx][0][:,0],boosting_model_outs[idx][0][:,1],\n", " c = y,cmap = cm_bright,edgecolors='white',s = 30,linewidths = 0.1)\n", " plt.axis('equal')\n", " plt.title(f'Outputs of {idx+1} classifiers')\n", " plt.colorbar()\n", " plt.savefig(f'Outputs of {idx+1} classifiers.png')\n", " plt.savefig(f'Outputs of {idx+1} classifiers.pdf')\n", " plt.show()\n", " \n", "#生成动图\n", "def create_gif(image_list, gif_name, duration=1):\n", " frames = []\n", " for image_name in image_list:\n", " frames.append(imageio.imread(image_name))\n", " imageio.mimsave(gif_name, frames, 'GIF', duration=duration)\n", " return\n", "\n", "def main():\n", " image_list = [f'Raw data ({n_samples} points).png']\n", " for i in range(len(boosting_model_outs)):\n", " image_list.append(f'Outputs of {i+1} classifiers.png')\n", " gif_name = '特征变换动图.gif'\n", " duration = 0.8\n", " create_gif(image_list, gif_name, duration)\n", "\n", "main()" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "O7ud1iEM_zFW" }, "outputs": [], "source": [] } ], "metadata": { "colab": { "provenance": [] }, "kernelspec": { "display_name": "Python 3", "name": "python3" }, "language_info": { "name": "python" } }, "nbformat": 4, "nbformat_minor": 0 }