keras_center_loss_with_embedding.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.datasets import mnist\n",
    "from keras.layers import Input,Conv2D,Activation,Dense,Lambda,Flatten,Embedding,PReLU,BatchNormalization\n",
    "from keras.models import Model\n",
    "import keras.backend as K\n",
    "from keras.utils import to_categorical\n",
    "import numpy as np\n",
    "from keras.callbacks import EarlyStopping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         (None, 28, 28, 1)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 26, 26, 32)        320       \n",
      "_________________________________________________________________\n",
      "batch_normalization_1 (Batch (None, 26, 26, 32)        128       \n",
      "_________________________________________________________________\n",
      "p_re_lu_1 (PReLU)            (None, 26, 26, 32)        21632     \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 24, 24, 32)        9248      \n",
      "_________________________________________________________________\n",
      "batch_normalization_2 (Batch (None, 24, 24, 32)        128       \n",
      "_________________________________________________________________\n",
      "p_re_lu_2 (PReLU)            (None, 24, 24, 32)        18432     \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 20, 20, 64)        51264     \n",
      "_________________________________________________________________\n",
      "batch_normalization_3 (Batch (None, 20, 20, 64)        256       \n",
      "_________________________________________________________________\n",
      "p_re_lu_3 (PReLU)            (None, 20, 20, 64)        25600     \n",
      "_________________________________________________________________\n",
      "conv2d_4 (Conv2D)            (None, 16, 16, 64)        102464    \n",
      "_________________________________________________________________\n",
      "batch_normalization_4 (Batch (None, 16, 16, 64)        256       \n",
      "_________________________________________________________________\n",
      "p_re_lu_4 (PReLU)            (None, 16, 16, 64)        16384     \n",
      "_________________________________________________________________\n",
      "conv2d_5 (Conv2D)            (None, 10, 10, 128)       401536    \n",
      "_________________________________________________________________\n",
      "batch_normalization_5 (Batch (None, 10, 10, 128)       512       \n",
      "_________________________________________________________________\n",
      "p_re_lu_5 (PReLU)            (None, 10, 10, 128)       12800     \n",
      "_________________________________________________________________\n",
      "conv2d_6 (Conv2D)            (None, 4, 4, 128)         802944    \n",
      "_________________________________________________________________\n",
      "batch_normalization_6 (Batch (None, 4, 4, 128)         512       \n",
      "_________________________________________________________________\n",
      "p_re_lu_6 (PReLU)            (None, 4, 4, 128)         2048      \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 2048)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 2)                 4098      \n",
      "_________________________________________________________________\n",
      "out1 (PReLU)                 (None, 2)                 2         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 10)                30        \n",
      "=================================================================\n",
      "Total params: 1,470,594\n",
      "Trainable params: 1,469,698\n",
      "Non-trainable params: 896\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "inputs=Input((28,28,1))\n",
    "x=Conv2D(32,(3,3))(inputs)\n",
    "x=BatchNormalization()(x)\n",
    "x=PReLU()(x)\n",
    "x=Conv2D(32,(3,3))(x)\n",
    "x=BatchNormalization()(x)\n",
    "x=PReLU()(x)\n",
    "x=Conv2D(64,(5,5))(x)\n",
    "x=BatchNormalization()(x)\n",
    "x=PReLU()(x)\n",
    "x=Conv2D(64,(5,5))(x)\n",
    "x=BatchNormalization()(x)\n",
    "x=PReLU()(x)\n",
    "x=Conv2D(128,(7,7))(x)\n",
    "x=BatchNormalization()(x)\n",
    "x=PReLU()(x)\n",
    "x=Conv2D(128,(7,7))(x)\n",
    "x=BatchNormalization()(x)\n",
    "x=PReLU()(x)\n",
    "x=Flatten()(x)\n",
    "x=Dense(2)(x)\n",
    "out1=PReLU(name=\"out1\")(x)\n",
    "out2=Dense(10,activation=\"softmax\")(out1)\n",
    "\n",
    "model=Model(inputs,out2)\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_1 (InputLayer)            (None, 28, 28, 1)    0                                            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_1 (Conv2D)               (None, 26, 26, 32)   320         input_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_1 (BatchNor (None, 26, 26, 32)   128         conv2d_1[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "p_re_lu_1 (PReLU)               (None, 26, 26, 32)   21632       batch_normalization_1[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_2 (Conv2D)               (None, 24, 24, 32)   9248        p_re_lu_1[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_2 (BatchNor (None, 24, 24, 32)   128         conv2d_2[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "p_re_lu_2 (PReLU)               (None, 24, 24, 32)   18432       batch_normalization_2[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_3 (Conv2D)               (None, 20, 20, 64)   51264       p_re_lu_2[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_3 (BatchNor (None, 20, 20, 64)   256         conv2d_3[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "p_re_lu_3 (PReLU)               (None, 20, 20, 64)   25600       batch_normalization_3[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_4 (Conv2D)               (None, 16, 16, 64)   102464      p_re_lu_3[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_4 (BatchNor (None, 16, 16, 64)   256         conv2d_4[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "p_re_lu_4 (PReLU)               (None, 16, 16, 64)   16384       batch_normalization_4[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_5 (Conv2D)               (None, 10, 10, 128)  401536      p_re_lu_4[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_5 (BatchNor (None, 10, 10, 128)  512         conv2d_5[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "p_re_lu_5 (PReLU)               (None, 10, 10, 128)  12800       batch_normalization_5[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_6 (Conv2D)               (None, 4, 4, 128)    802944      p_re_lu_5[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_6 (BatchNor (None, 4, 4, 128)    512         conv2d_6[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "p_re_lu_6 (PReLU)               (None, 4, 4, 128)    2048        batch_normalization_6[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "flatten_1 (Flatten)             (None, 2048)         0           p_re_lu_6[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "dense_1 (Dense)                 (None, 2)            4098        flatten_1[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "input_2 (InputLayer)            (None, 1)            0                                            \n",
      "__________________________________________________________________________________________________\n",
      "out1 (PReLU)                    (None, 2)            2           dense_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "embedding_1 (Embedding)         (None, 1, 2)         20          input_2[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "dense_2 (Dense)                 (None, 10)           30          out1[0][0]                       \n",
      "__________________________________________________________________________________________________\n",
      "lambda_1 (Lambda)               (None, 1)            0           out1[0][0]                       \n",
      "                                                                 embedding_1[0][0]                \n",
      "==================================================================================================\n",
      "Total params: 1,470,614\n",
      "Trainable params: 1,469,718\n",
      "Non-trainable params: 896\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "lambda_c=1\n",
    "input_=Input(shape=(1,))\n",
    "centers=Embedding(10,2)(input_)\n",
    "intra_loss=Lambda(lambda x:K.sum(K.square(x[0]-x[1][:,0]),1,keepdims=True))([out1,centers])\n",
    "model_center_loss=Model([inputs,input_],[out2,intra_loss])\n",
    "model_center_loss.compile(optimizer=\"sgd\",\n",
    "                          loss=[\"categorical_crossentropy\",lambda y_true,y_pred:y_pred],\n",
    "                          loss_weights=[1,lambda_c/2.],\n",
    "                          metrics=[\"acc\"])\n",
    "model_center_loss.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(30000, 28, 28, 1)\n",
      "(30000,)\n",
      "(30000, 10)\n"
     ]
    }
   ],
   "source": [
    "path=\"mnist.npz\"\n",
    "f = np.load(path)\n",
    "x_train, y_train = f['x_train'][:30000], f['y_train'][:30000]\n",
    "x_test, y_test = f['x_test'], f['y_test']\n",
    "f.close()\n",
    "\n",
    "x_train=x_train.astype(\"float32\")\n",
    "x_test=x_test.astype(\"float32\")\n",
    "x_train/=255.\n",
    "x_test/=255.\n",
    "x_train=np.expand_dims(x_train,axis=-1)\n",
    "x_test=np.expand_dims(x_test,axis=-1)\n",
    "\n",
    "y_train_one_hot=to_categorical(y_train)\n",
    "y_test_one_hot=to_categorical(y_test)\n",
    "\n",
    "print(x_train.shape)\n",
    "print(y_train.shape)\n",
    "print(y_train_one_hot.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 30000 samples, validate on 10000 samples\n",
      "Epoch 1/100\n",
      "30000/30000 [==============================] - 7s 235us/step - loss: 2.3248 - dense_2_loss: 2.3028 - lambda_1_loss: 0.0439 - dense_2_acc: 0.1428 - lambda_1_acc: 0.9852 - val_loss: 2.3019 - val_dense_2_loss: 2.2986 - val_lambda_1_loss: 0.0067 - val_dense_2_acc: 0.1694 - val_lambda_1_acc: 0.9982\n",
      "Epoch 2/100\n",
      "30000/30000 [==============================] - 5s 158us/step - loss: 2.3004 - dense_2_loss: 2.2945 - lambda_1_loss: 0.0117 - dense_2_acc: 0.1817 - lambda_1_acc: 0.9940 - val_loss: 2.2989 - val_dense_2_loss: 2.2895 - val_lambda_1_loss: 0.0188 - val_dense_2_acc: 0.1922 - val_lambda_1_acc: 0.9909\n",
      "Epoch 3/100\n",
      "30000/30000 [==============================] - 5s 163us/step - loss: 2.2962 - dense_2_loss: 2.2851 - lambda_1_loss: 0.0222 - dense_2_acc: 0.1953 - lambda_1_acc: 0.9895 - val_loss: 2.2947 - val_dense_2_loss: 2.2820 - val_lambda_1_loss: 0.0255 - val_dense_2_acc: 0.1949 - val_lambda_1_acc: 0.9899\n",
      "Epoch 4/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 2.2900 - dense_2_loss: 2.2722 - lambda_1_loss: 0.0356 - dense_2_acc: 0.2042 - lambda_1_acc: 0.9832 - val_loss: 2.2897 - val_dense_2_loss: 2.2717 - val_lambda_1_loss: 0.0360 - val_dense_2_acc: 0.2042 - val_lambda_1_acc: 0.9876\n",
      "Epoch 5/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 2.2828 - dense_2_loss: 2.2585 - lambda_1_loss: 0.0486 - dense_2_acc: 0.2082 - lambda_1_acc: 0.9789 - val_loss: 2.2849 - val_dense_2_loss: 2.2492 - val_lambda_1_loss: 0.0714 - val_dense_2_acc: 0.2170 - val_lambda_1_acc: 0.9713\n",
      "Epoch 6/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 2.2754 - dense_2_loss: 2.2458 - lambda_1_loss: 0.0592 - dense_2_acc: 0.2119 - lambda_1_acc: 0.9738 - val_loss: 2.2781 - val_dense_2_loss: 2.2447 - val_lambda_1_loss: 0.0670 - val_dense_2_acc: 0.2163 - val_lambda_1_acc: 0.9749\n",
      "Epoch 7/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 2.2677 - dense_2_loss: 2.2338 - lambda_1_loss: 0.0678 - dense_2_acc: 0.2135 - lambda_1_acc: 0.9707 - val_loss: 2.2719 - val_dense_2_loss: 2.2297 - val_lambda_1_loss: 0.0843 - val_dense_2_acc: 0.2131 - val_lambda_1_acc: 0.9669\n",
      "Epoch 8/100\n",
      "30000/30000 [==============================] - 5s 159us/step - loss: 2.2596 - dense_2_loss: 2.2216 - lambda_1_loss: 0.0760 - dense_2_acc: 0.2149 - lambda_1_acc: 0.9665 - val_loss: 2.2654 - val_dense_2_loss: 2.2165 - val_lambda_1_loss: 0.0977 - val_dense_2_acc: 0.2159 - val_lambda_1_acc: 0.9601\n",
      "Epoch 9/100\n",
      "30000/30000 [==============================] - 5s 159us/step - loss: 2.2505 - dense_2_loss: 2.2090 - lambda_1_loss: 0.0831 - dense_2_acc: 0.2118 - lambda_1_acc: 0.9633 - val_loss: 2.2573 - val_dense_2_loss: 2.2044 - val_lambda_1_loss: 0.1060 - val_dense_2_acc: 0.2009 - val_lambda_1_acc: 0.9566\n",
      "Epoch 10/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 2.2411 - dense_2_loss: 2.1956 - lambda_1_loss: 0.0910 - dense_2_acc: 0.2106 - lambda_1_acc: 0.9582 - val_loss: 2.2499 - val_dense_2_loss: 2.1933 - val_lambda_1_loss: 0.1132 - val_dense_2_acc: 0.2282 - val_lambda_1_acc: 0.9514\n",
      "Epoch 11/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 2.2302 - dense_2_loss: 2.1814 - lambda_1_loss: 0.0976 - dense_2_acc: 0.2085 - lambda_1_acc: 0.9544 - val_loss: 2.2380 - val_dense_2_loss: 2.1780 - val_lambda_1_loss: 0.1200 - val_dense_2_acc: 0.2057 - val_lambda_1_acc: 0.9486\n",
      "Epoch 12/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 2.2185 - dense_2_loss: 2.1660 - lambda_1_loss: 0.1050 - dense_2_acc: 0.2071 - lambda_1_acc: 0.9499 - val_loss: 2.2260 - val_dense_2_loss: 2.1586 - val_lambda_1_loss: 0.1348 - val_dense_2_acc: 0.2002 - val_lambda_1_acc: 0.9417\n",
      "Epoch 13/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 2.2049 - dense_2_loss: 2.1461 - lambda_1_loss: 0.1175 - dense_2_acc: 0.2093 - lambda_1_acc: 0.9428 - val_loss: 2.2125 - val_dense_2_loss: 2.1244 - val_lambda_1_loss: 0.1763 - val_dense_2_acc: 0.2116 - val_lambda_1_acc: 0.9209\n",
      "Epoch 14/100\n",
      "30000/30000 [==============================] - 5s 162us/step - loss: 2.1860 - dense_2_loss: 2.1139 - lambda_1_loss: 0.1442 - dense_2_acc: 0.2245 - lambda_1_acc: 0.9262 - val_loss: 2.1922 - val_dense_2_loss: 2.1009 - val_lambda_1_loss: 0.1825 - val_dense_2_acc: 0.2433 - val_lambda_1_acc: 0.9135\n",
      "Epoch 15/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 2.1616 - dense_2_loss: 2.0722 - lambda_1_loss: 0.1789 - dense_2_acc: 0.2592 - lambda_1_acc: 0.9027 - val_loss: 2.1724 - val_dense_2_loss: 2.0343 - val_lambda_1_loss: 0.2761 - val_dense_2_acc: 0.2726 - val_lambda_1_acc: 0.8509\n",
      "Epoch 16/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 2.1320 - dense_2_loss: 2.0288 - lambda_1_loss: 0.2063 - dense_2_acc: 0.2909 - lambda_1_acc: 0.8802 - val_loss: 2.1468 - val_dense_2_loss: 2.0197 - val_lambda_1_loss: 0.2542 - val_dense_2_acc: 0.3037 - val_lambda_1_acc: 0.8646\n",
      "Epoch 17/100\n",
      "30000/30000 [==============================] - 5s 158us/step - loss: 2.1036 - dense_2_loss: 1.9925 - lambda_1_loss: 0.2221 - dense_2_acc: 0.3137 - lambda_1_acc: 0.8639 - val_loss: 2.1245 - val_dense_2_loss: 1.9996 - val_lambda_1_loss: 0.2498 - val_dense_2_acc: 0.3090 - val_lambda_1_acc: 0.8672\n",
      "Epoch 18/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 2.0818 - dense_2_loss: 1.9622 - lambda_1_loss: 0.2391 - dense_2_acc: 0.3311 - lambda_1_acc: 0.8556 - val_loss: 2.1131 - val_dense_2_loss: 1.9231 - val_lambda_1_loss: 0.3800 - val_dense_2_acc: 0.3131 - val_lambda_1_acc: 0.7746\n",
      "Epoch 19/100\n",
      "30000/30000 [==============================] - 5s 159us/step - loss: 2.0501 - dense_2_loss: 1.9301 - lambda_1_loss: 0.2402 - dense_2_acc: 0.3421 - lambda_1_acc: 0.8481 - val_loss: 2.0963 - val_dense_2_loss: 1.9144 - val_lambda_1_loss: 0.3639 - val_dense_2_acc: 0.3405 - val_lambda_1_acc: 0.8019\n",
      "Epoch 20/100\n",
      "30000/30000 [==============================] - 5s 159us/step - loss: 2.0238 - dense_2_loss: 1.9007 - lambda_1_loss: 0.2463 - dense_2_acc: 0.3512 - lambda_1_acc: 0.8416 - val_loss: 2.0477 - val_dense_2_loss: 1.9065 - val_lambda_1_loss: 0.2825 - val_dense_2_acc: 0.3500 - val_lambda_1_acc: 0.8433\n",
      "Epoch 21/100\n",
      "30000/30000 [==============================] - 5s 162us/step - loss: 1.9963 - dense_2_loss: 1.8712 - lambda_1_loss: 0.2501 - dense_2_acc: 0.3660 - lambda_1_acc: 0.8360 - val_loss: 2.0233 - val_dense_2_loss: 1.8888 - val_lambda_1_loss: 0.2690 - val_dense_2_acc: 0.3502 - val_lambda_1_acc: 0.8382\n",
      "Epoch 22/100\n",
      "30000/30000 [==============================] - 5s 162us/step - loss: 1.9676 - dense_2_loss: 1.8426 - lambda_1_loss: 0.2500 - dense_2_acc: 0.3800 - lambda_1_acc: 0.8349 - val_loss: 1.9956 - val_dense_2_loss: 1.8573 - val_lambda_1_loss: 0.2766 - val_dense_2_acc: 0.3780 - val_lambda_1_acc: 0.8355\n",
      "Epoch 23/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 1.9412 - dense_2_loss: 1.8144 - lambda_1_loss: 0.2536 - dense_2_acc: 0.3933 - lambda_1_acc: 0.8296 - val_loss: 1.9794 - val_dense_2_loss: 1.8347 - val_lambda_1_loss: 0.2894 - val_dense_2_acc: 0.3940 - val_lambda_1_acc: 0.8281\n",
      "Epoch 24/100\n",
      "30000/30000 [==============================] - 5s 163us/step - loss: 1.9202 - dense_2_loss: 1.7892 - lambda_1_loss: 0.2621 - dense_2_acc: 0.4065 - lambda_1_acc: 0.8283 - val_loss: 1.9648 - val_dense_2_loss: 1.8082 - val_lambda_1_loss: 0.3132 - val_dense_2_acc: 0.3921 - val_lambda_1_acc: 0.8187\n",
      "Epoch 25/100\n",
      "30000/30000 [==============================] - 5s 170us/step - loss: 1.8849 - dense_2_loss: 1.7582 - lambda_1_loss: 0.2533 - dense_2_acc: 0.4184 - lambda_1_acc: 0.8340 - val_loss: 1.9260 - val_dense_2_loss: 1.7743 - val_lambda_1_loss: 0.3034 - val_dense_2_acc: 0.4148 - val_lambda_1_acc: 0.8265\n",
      "Epoch 26/100\n",
      "30000/30000 [==============================] - 5s 163us/step - loss: 1.8605 - dense_2_loss: 1.7331 - lambda_1_loss: 0.2548 - dense_2_acc: 0.4268 - lambda_1_acc: 0.8342 - val_loss: 1.9012 - val_dense_2_loss: 1.7812 - val_lambda_1_loss: 0.2400 - val_dense_2_acc: 0.4100 - val_lambda_1_acc: 0.8627\n",
      "Epoch 27/100\n",
      "30000/30000 [==============================] - 5s 170us/step - loss: 1.8303 - dense_2_loss: 1.7044 - lambda_1_loss: 0.2516 - dense_2_acc: 0.4348 - lambda_1_acc: 0.8381 - val_loss: 1.8806 - val_dense_2_loss: 1.7333 - val_lambda_1_loss: 0.2946 - val_dense_2_acc: 0.4137 - val_lambda_1_acc: 0.8231\n",
      "Epoch 28/100\n",
      "30000/30000 [==============================] - 5s 166us/step - loss: 1.8037 - dense_2_loss: 1.6788 - lambda_1_loss: 0.2499 - dense_2_acc: 0.4411 - lambda_1_acc: 0.8442 - val_loss: 1.8668 - val_dense_2_loss: 1.7460 - val_lambda_1_loss: 0.2416 - val_dense_2_acc: 0.4226 - val_lambda_1_acc: 0.8618\n",
      "Epoch 29/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 1.7782 - dense_2_loss: 1.6536 - lambda_1_loss: 0.2492 - dense_2_acc: 0.4472 - lambda_1_acc: 0.8462 - val_loss: 1.8245 - val_dense_2_loss: 1.6856 - val_lambda_1_loss: 0.2779 - val_dense_2_acc: 0.4272 - val_lambda_1_acc: 0.8342\n",
      "Epoch 30/100\n",
      "30000/30000 [==============================] - 5s 164us/step - loss: 1.7517 - dense_2_loss: 1.6292 - lambda_1_loss: 0.2450 - dense_2_acc: 0.4506 - lambda_1_acc: 0.8530 - val_loss: 1.8111 - val_dense_2_loss: 1.6465 - val_lambda_1_loss: 0.3291 - val_dense_2_acc: 0.4267 - val_lambda_1_acc: 0.7973\n",
      "Epoch 31/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 1.7334 - dense_2_loss: 1.6080 - lambda_1_loss: 0.2508 - dense_2_acc: 0.4546 - lambda_1_acc: 0.8454 - val_loss: 1.7932 - val_dense_2_loss: 1.6320 - val_lambda_1_loss: 0.3224 - val_dense_2_acc: 0.4280 - val_lambda_1_acc: 0.8056\n",
      "Epoch 32/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 1.7028 - dense_2_loss: 1.5819 - lambda_1_loss: 0.2417 - dense_2_acc: 0.4568 - lambda_1_acc: 0.8619 - val_loss: 1.7644 - val_dense_2_loss: 1.6183 - val_lambda_1_loss: 0.2922 - val_dense_2_acc: 0.4331 - val_lambda_1_acc: 0.8304\n",
      "Epoch 33/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 1.6722 - dense_2_loss: 1.5563 - lambda_1_loss: 0.2318 - dense_2_acc: 0.4622 - lambda_1_acc: 0.8762 - val_loss: 1.7362 - val_dense_2_loss: 1.5943 - val_lambda_1_loss: 0.2838 - val_dense_2_acc: 0.4433 - val_lambda_1_acc: 0.8308\n",
      "Epoch 34/100\n",
      "30000/30000 [==============================] - 5s 169us/step - loss: 1.6526 - dense_2_loss: 1.5355 - lambda_1_loss: 0.2342 - dense_2_acc: 0.4660 - lambda_1_acc: 0.8742 - val_loss: 1.7169 - val_dense_2_loss: 1.5858 - val_lambda_1_loss: 0.2622 - val_dense_2_acc: 0.4523 - val_lambda_1_acc: 0.8460\n",
      "Epoch 35/100\n",
      "30000/30000 [==============================] - 5s 167us/step - loss: 1.6302 - dense_2_loss: 1.5146 - lambda_1_loss: 0.2311 - dense_2_acc: 0.4707 - lambda_1_acc: 0.8792 - val_loss: 1.7156 - val_dense_2_loss: 1.5604 - val_lambda_1_loss: 0.3105 - val_dense_2_acc: 0.4414 - val_lambda_1_acc: 0.8168\n",
      "Epoch 36/100\n",
      "30000/30000 [==============================] - 5s 162us/step - loss: 1.6037 - dense_2_loss: 1.4914 - lambda_1_loss: 0.2245 - dense_2_acc: 0.4709 - lambda_1_acc: 0.8875 - val_loss: 1.6909 - val_dense_2_loss: 1.5324 - val_lambda_1_loss: 0.3171 - val_dense_2_acc: 0.4530 - val_lambda_1_acc: 0.8061\n",
      "Epoch 37/100\n",
      "30000/30000 [==============================] - 5s 163us/step - loss: 1.5835 - dense_2_loss: 1.4713 - lambda_1_loss: 0.2244 - dense_2_acc: 0.4774 - lambda_1_acc: 0.8849 - val_loss: 1.6732 - val_dense_2_loss: 1.5340 - val_lambda_1_loss: 0.2785 - val_dense_2_acc: 0.4600 - val_lambda_1_acc: 0.8291\n",
      "Epoch 38/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 1.5615 - dense_2_loss: 1.4512 - lambda_1_loss: 0.2207 - dense_2_acc: 0.4826 - lambda_1_acc: 0.8924 - val_loss: 1.6706 - val_dense_2_loss: 1.4840 - val_lambda_1_loss: 0.3732 - val_dense_2_acc: 0.4550 - val_lambda_1_acc: 0.7782\n",
      "Epoch 39/100\n",
      "30000/30000 [==============================] - 5s 162us/step - loss: 1.5350 - dense_2_loss: 1.4282 - lambda_1_loss: 0.2137 - dense_2_acc: 0.4848 - lambda_1_acc: 0.9036 - val_loss: 1.6189 - val_dense_2_loss: 1.5013 - val_lambda_1_loss: 0.2353 - val_dense_2_acc: 0.4648 - val_lambda_1_acc: 0.8713\n",
      "Epoch 40/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 1.5164 - dense_2_loss: 1.4099 - lambda_1_loss: 0.2131 - dense_2_acc: 0.4950 - lambda_1_acc: 0.9045 - val_loss: 1.5927 - val_dense_2_loss: 1.4770 - val_lambda_1_loss: 0.2314 - val_dense_2_acc: 0.4698 - val_lambda_1_acc: 0.8748\n",
      "Epoch 41/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 1.4950 - dense_2_loss: 1.3903 - lambda_1_loss: 0.2094 - dense_2_acc: 0.5010 - lambda_1_acc: 0.9068 - val_loss: 1.5787 - val_dense_2_loss: 1.4298 - val_lambda_1_loss: 0.2977 - val_dense_2_acc: 0.4775 - val_lambda_1_acc: 0.8242\n",
      "Epoch 42/100\n",
      "30000/30000 [==============================] - 5s 164us/step - loss: 1.4779 - dense_2_loss: 1.3724 - lambda_1_loss: 0.2112 - dense_2_acc: 0.5081 - lambda_1_acc: 0.8980 - val_loss: 1.5660 - val_dense_2_loss: 1.4629 - val_lambda_1_loss: 0.2063 - val_dense_2_acc: 0.5226 - val_lambda_1_acc: 0.9004\n",
      "Epoch 43/100\n",
      "30000/30000 [==============================] - 5s 162us/step - loss: 1.4501 - dense_2_loss: 1.3503 - lambda_1_loss: 0.1996 - dense_2_acc: 0.5237 - lambda_1_acc: 0.9198 - val_loss: 1.5428 - val_dense_2_loss: 1.4052 - val_lambda_1_loss: 0.2752 - val_dense_2_acc: 0.4914 - val_lambda_1_acc: 0.8449\n",
      "Epoch 44/100\n",
      "30000/30000 [==============================] - 5s 164us/step - loss: 1.4320 - dense_2_loss: 1.3317 - lambda_1_loss: 0.2007 - dense_2_acc: 0.5320 - lambda_1_acc: 0.9166 - val_loss: 1.5267 - val_dense_2_loss: 1.3967 - val_lambda_1_loss: 0.2599 - val_dense_2_acc: 0.5430 - val_lambda_1_acc: 0.8564\n",
      "Epoch 45/100\n",
      "30000/30000 [==============================] - 5s 164us/step - loss: 1.4154 - dense_2_loss: 1.3145 - lambda_1_loss: 0.2018 - dense_2_acc: 0.5509 - lambda_1_acc: 0.9153 - val_loss: 1.5001 - val_dense_2_loss: 1.3744 - val_lambda_1_loss: 0.2513 - val_dense_2_acc: 0.5570 - val_lambda_1_acc: 0.8595\n",
      "Epoch 46/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 1.3972 - dense_2_loss: 1.2975 - lambda_1_loss: 0.1993 - dense_2_acc: 0.5618 - lambda_1_acc: 0.9184 - val_loss: 1.4860 - val_dense_2_loss: 1.3388 - val_lambda_1_loss: 0.2944 - val_dense_2_acc: 0.5421 - val_lambda_1_acc: 0.8210\n",
      "Epoch 47/100\n",
      "30000/30000 [==============================] - 5s 163us/step - loss: 1.3702 - dense_2_loss: 1.2744 - lambda_1_loss: 0.1916 - dense_2_acc: 0.5796 - lambda_1_acc: 0.9314 - val_loss: 1.4863 - val_dense_2_loss: 1.3798 - val_lambda_1_loss: 0.2129 - val_dense_2_acc: 0.5724 - val_lambda_1_acc: 0.9005\n",
      "Epoch 48/100\n",
      "30000/30000 [==============================] - 5s 162us/step - loss: 1.3596 - dense_2_loss: 1.2611 - lambda_1_loss: 0.1970 - dense_2_acc: 0.5890 - lambda_1_acc: 0.9208 - val_loss: 1.4583 - val_dense_2_loss: 1.3260 - val_lambda_1_loss: 0.2647 - val_dense_2_acc: 0.5834 - val_lambda_1_acc: 0.8524\n",
      "Epoch 49/100\n",
      "30000/30000 [==============================] - 5s 158us/step - loss: 1.3345 - dense_2_loss: 1.2404 - lambda_1_loss: 0.1882 - dense_2_acc: 0.6009 - lambda_1_acc: 0.9326 - val_loss: 1.4338 - val_dense_2_loss: 1.3008 - val_lambda_1_loss: 0.2662 - val_dense_2_acc: 0.5904 - val_lambda_1_acc: 0.8486\n",
      "Epoch 50/100\n",
      "30000/30000 [==============================] - 5s 163us/step - loss: 1.3188 - dense_2_loss: 1.2233 - lambda_1_loss: 0.1911 - dense_2_acc: 0.6074 - lambda_1_acc: 0.9300 - val_loss: 1.4692 - val_dense_2_loss: 1.2915 - val_lambda_1_loss: 0.3553 - val_dense_2_acc: 0.6036 - val_lambda_1_acc: 0.7877\n",
      "Epoch 51/100\n",
      "30000/30000 [==============================] - 5s 163us/step - loss: 1.3116 - dense_2_loss: 1.2115 - lambda_1_loss: 0.2002 - dense_2_acc: 0.6113 - lambda_1_acc: 0.9147 - val_loss: 1.4014 - val_dense_2_loss: 1.2822 - val_lambda_1_loss: 0.2384 - val_dense_2_acc: 0.5988 - val_lambda_1_acc: 0.8812\n",
      "Epoch 52/100\n",
      "30000/30000 [==============================] - 5s 164us/step - loss: 1.2756 - dense_2_loss: 1.1855 - lambda_1_loss: 0.1803 - dense_2_acc: 0.6207 - lambda_1_acc: 0.9453 - val_loss: 1.4018 - val_dense_2_loss: 1.2895 - val_lambda_1_loss: 0.2247 - val_dense_2_acc: 0.6053 - val_lambda_1_acc: 0.8983\n",
      "Epoch 53/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 1.2732 - dense_2_loss: 1.1762 - lambda_1_loss: 0.1939 - dense_2_acc: 0.6274 - lambda_1_acc: 0.9236 - val_loss: 1.3797 - val_dense_2_loss: 1.2301 - val_lambda_1_loss: 0.2993 - val_dense_2_acc: 0.6282 - val_lambda_1_acc: 0.8230\n",
      "Epoch 54/100\n",
      "30000/30000 [==============================] - 5s 159us/step - loss: 1.2607 - dense_2_loss: 1.1614 - lambda_1_loss: 0.1986 - dense_2_acc: 0.6336 - lambda_1_acc: 0.9158 - val_loss: 1.3572 - val_dense_2_loss: 1.2374 - val_lambda_1_loss: 0.2396 - val_dense_2_acc: 0.6133 - val_lambda_1_acc: 0.8795\n",
      "Epoch 55/100\n",
      "30000/30000 [==============================] - 5s 162us/step - loss: 1.2354 - dense_2_loss: 1.1417 - lambda_1_loss: 0.1873 - dense_2_acc: 0.6395 - lambda_1_acc: 0.9281 - val_loss: 1.3434 - val_dense_2_loss: 1.2089 - val_lambda_1_loss: 0.2690 - val_dense_2_acc: 0.6169 - val_lambda_1_acc: 0.8559\n",
      "Epoch 56/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 1.2222 - dense_2_loss: 1.1255 - lambda_1_loss: 0.1934 - dense_2_acc: 0.6465 - lambda_1_acc: 0.9264 - val_loss: 1.3239 - val_dense_2_loss: 1.1974 - val_lambda_1_loss: 0.2530 - val_dense_2_acc: 0.6311 - val_lambda_1_acc: 0.8755\n",
      "Epoch 57/100\n",
      "30000/30000 [==============================] - 5s 165us/step - loss: 1.1979 - dense_2_loss: 1.1069 - lambda_1_loss: 0.1820 - dense_2_acc: 0.6495 - lambda_1_acc: 0.9392 - val_loss: 1.3053 - val_dense_2_loss: 1.1787 - val_lambda_1_loss: 0.2532 - val_dense_2_acc: 0.6379 - val_lambda_1_acc: 0.8716\n",
      "Epoch 58/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 1.1720 - dense_2_loss: 1.0860 - lambda_1_loss: 0.1719 - dense_2_acc: 0.6550 - lambda_1_acc: 0.9497 - val_loss: 1.2921 - val_dense_2_loss: 1.1673 - val_lambda_1_loss: 0.2497 - val_dense_2_acc: 0.6408 - val_lambda_1_acc: 0.8755\n",
      "Epoch 59/100\n",
      "30000/30000 [==============================] - 5s 167us/step - loss: 1.1547 - dense_2_loss: 1.0688 - lambda_1_loss: 0.1718 - dense_2_acc: 0.6597 - lambda_1_acc: 0.9526 - val_loss: 1.2894 - val_dense_2_loss: 1.1742 - val_lambda_1_loss: 0.2305 - val_dense_2_acc: 0.6442 - val_lambda_1_acc: 0.8969\n",
      "Epoch 60/100\n",
      "30000/30000 [==============================] - 5s 163us/step - loss: 1.1479 - dense_2_loss: 1.0586 - lambda_1_loss: 0.1786 - dense_2_acc: 0.6725 - lambda_1_acc: 0.9433 - val_loss: 1.2864 - val_dense_2_loss: 1.1514 - val_lambda_1_loss: 0.2700 - val_dense_2_acc: 0.6662 - val_lambda_1_acc: 0.8637\n",
      "Epoch 61/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 1.1299 - dense_2_loss: 1.0421 - lambda_1_loss: 0.1756 - dense_2_acc: 0.6776 - lambda_1_acc: 0.9455 - val_loss: 1.2737 - val_dense_2_loss: 1.1315 - val_lambda_1_loss: 0.2844 - val_dense_2_acc: 0.6560 - val_lambda_1_acc: 0.8515\n",
      "Epoch 62/100\n",
      "30000/30000 [==============================] - 5s 162us/step - loss: 1.1236 - dense_2_loss: 1.0311 - lambda_1_loss: 0.1849 - dense_2_acc: 0.6808 - lambda_1_acc: 0.9329 - val_loss: 1.2570 - val_dense_2_loss: 1.1098 - val_lambda_1_loss: 0.2945 - val_dense_2_acc: 0.6540 - val_lambda_1_acc: 0.8375\n",
      "Epoch 63/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 1.0999 - dense_2_loss: 1.0123 - lambda_1_loss: 0.1753 - dense_2_acc: 0.6926 - lambda_1_acc: 0.9414 - val_loss: 1.2243 - val_dense_2_loss: 1.0814 - val_lambda_1_loss: 0.2858 - val_dense_2_acc: 0.6693 - val_lambda_1_acc: 0.8523\n",
      "Epoch 64/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 1.0937 - dense_2_loss: 1.0020 - lambda_1_loss: 0.1834 - dense_2_acc: 0.7091 - lambda_1_acc: 0.9329 - val_loss: 1.2113 - val_dense_2_loss: 1.0834 - val_lambda_1_loss: 0.2558 - val_dense_2_acc: 0.6753 - val_lambda_1_acc: 0.8811\n",
      "Epoch 65/100\n",
      "30000/30000 [==============================] - 5s 165us/step - loss: 1.0664 - dense_2_loss: 0.9815 - lambda_1_loss: 0.1698 - dense_2_acc: 0.7097 - lambda_1_acc: 0.9492 - val_loss: 1.5035 - val_dense_2_loss: 1.2428 - val_lambda_1_loss: 0.5213 - val_dense_2_acc: 0.5690 - val_lambda_1_acc: 0.6462\n",
      "Epoch 66/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 1.0694 - dense_2_loss: 0.9778 - lambda_1_loss: 0.1833 - dense_2_acc: 0.7117 - lambda_1_acc: 0.9374 - val_loss: 1.1952 - val_dense_2_loss: 1.0492 - val_lambda_1_loss: 0.2919 - val_dense_2_acc: 0.7095 - val_lambda_1_acc: 0.8476\n",
      "Epoch 67/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 1.0394 - dense_2_loss: 0.9549 - lambda_1_loss: 0.1690 - dense_2_acc: 0.7196 - lambda_1_acc: 0.9481 - val_loss: 1.2066 - val_dense_2_loss: 1.0761 - val_lambda_1_loss: 0.2610 - val_dense_2_acc: 0.6787 - val_lambda_1_acc: 0.8776\n",
      "Epoch 68/100\n",
      "30000/30000 [==============================] - 5s 162us/step - loss: 1.0388 - dense_2_loss: 0.9481 - lambda_1_loss: 0.1814 - dense_2_acc: 0.7307 - lambda_1_acc: 0.9362 - val_loss: 1.1583 - val_dense_2_loss: 1.0302 - val_lambda_1_loss: 0.2564 - val_dense_2_acc: 0.7254 - val_lambda_1_acc: 0.8747\n",
      "Epoch 69/100\n",
      "30000/30000 [==============================] - 5s 164us/step - loss: 1.0078 - dense_2_loss: 0.9270 - lambda_1_loss: 0.1615 - dense_2_acc: 0.7464 - lambda_1_acc: 0.9529 - val_loss: 1.1547 - val_dense_2_loss: 1.0182 - val_lambda_1_loss: 0.2728 - val_dense_2_acc: 0.7122 - val_lambda_1_acc: 0.8701\n",
      "Epoch 70/100\n",
      "30000/30000 [==============================] - 5s 164us/step - loss: 0.9985 - dense_2_loss: 0.9154 - lambda_1_loss: 0.1662 - dense_2_acc: 0.7541 - lambda_1_acc: 0.9491 - val_loss: 1.1836 - val_dense_2_loss: 1.0483 - val_lambda_1_loss: 0.2707 - val_dense_2_acc: 0.7294 - val_lambda_1_acc: 0.8556\n",
      "Epoch 71/100\n",
      "30000/30000 [==============================] - 5s 163us/step - loss: 0.9777 - dense_2_loss: 0.8998 - lambda_1_loss: 0.1558 - dense_2_acc: 0.7608 - lambda_1_acc: 0.9605 - val_loss: 1.1270 - val_dense_2_loss: 0.9933 - val_lambda_1_loss: 0.2674 - val_dense_2_acc: 0.7412 - val_lambda_1_acc: 0.8689\n",
      "Epoch 72/100\n",
      "30000/30000 [==============================] - 5s 164us/step - loss: 0.9786 - dense_2_loss: 0.8927 - lambda_1_loss: 0.1718 - dense_2_acc: 0.7707 - lambda_1_acc: 0.9445 - val_loss: 1.1312 - val_dense_2_loss: 0.9808 - val_lambda_1_loss: 0.3007 - val_dense_2_acc: 0.7536 - val_lambda_1_acc: 0.8301\n",
      "Epoch 73/100\n",
      "30000/30000 [==============================] - 5s 164us/step - loss: 0.9540 - dense_2_loss: 0.8765 - lambda_1_loss: 0.1549 - dense_2_acc: 0.7823 - lambda_1_acc: 0.9599 - val_loss: 1.1271 - val_dense_2_loss: 0.9731 - val_lambda_1_loss: 0.3080 - val_dense_2_acc: 0.7164 - val_lambda_1_acc: 0.8395\n",
      "Epoch 74/100\n",
      "30000/30000 [==============================] - 5s 162us/step - loss: 0.9505 - dense_2_loss: 0.8677 - lambda_1_loss: 0.1656 - dense_2_acc: 0.7914 - lambda_1_acc: 0.9481 - val_loss: 1.0995 - val_dense_2_loss: 0.9717 - val_lambda_1_loss: 0.2555 - val_dense_2_acc: 0.7512 - val_lambda_1_acc: 0.8885\n",
      "Epoch 75/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 0.9376 - dense_2_loss: 0.8558 - lambda_1_loss: 0.1636 - dense_2_acc: 0.7976 - lambda_1_acc: 0.9517 - val_loss: 1.1255 - val_dense_2_loss: 0.9787 - val_lambda_1_loss: 0.2934 - val_dense_2_acc: 0.7232 - val_lambda_1_acc: 0.8509\n",
      "Epoch 76/100\n",
      "30000/30000 [==============================] - 5s 163us/step - loss: 0.9166 - dense_2_loss: 0.8410 - lambda_1_loss: 0.1512 - dense_2_acc: 0.8155 - lambda_1_acc: 0.9601 - val_loss: 1.1189 - val_dense_2_loss: 0.9551 - val_lambda_1_loss: 0.3277 - val_dense_2_acc: 0.7345 - val_lambda_1_acc: 0.8109\n",
      "Epoch 77/100\n",
      "30000/30000 [==============================] - 5s 163us/step - loss: 0.9064 - dense_2_loss: 0.8293 - lambda_1_loss: 0.1542 - dense_2_acc: 0.8277 - lambda_1_acc: 0.9582 - val_loss: 1.1283 - val_dense_2_loss: 0.9616 - val_lambda_1_loss: 0.3334 - val_dense_2_acc: 0.7304 - val_lambda_1_acc: 0.8260\n",
      "Epoch 78/100\n",
      "30000/30000 [==============================] - 5s 166us/step - loss: 0.9004 - dense_2_loss: 0.8210 - lambda_1_loss: 0.1589 - dense_2_acc: 0.8355 - lambda_1_acc: 0.9527 - val_loss: 1.0665 - val_dense_2_loss: 0.9258 - val_lambda_1_loss: 0.2814 - val_dense_2_acc: 0.7850 - val_lambda_1_acc: 0.8647\n",
      "Epoch 79/100\n",
      "30000/30000 [==============================] - 5s 162us/step - loss: 0.8813 - dense_2_loss: 0.8068 - lambda_1_loss: 0.1490 - dense_2_acc: 0.8500 - lambda_1_acc: 0.9610 - val_loss: 1.0752 - val_dense_2_loss: 0.9301 - val_lambda_1_loss: 0.2901 - val_dense_2_acc: 0.7628 - val_lambda_1_acc: 0.8683\n",
      "Epoch 80/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 0.8638 - dense_2_loss: 0.7930 - lambda_1_loss: 0.1416 - dense_2_acc: 0.8651 - lambda_1_acc: 0.9675 - val_loss: 1.0630 - val_dense_2_loss: 0.9055 - val_lambda_1_loss: 0.3150 - val_dense_2_acc: 0.8033 - val_lambda_1_acc: 0.8317\n",
      "Epoch 81/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 0.8501 - dense_2_loss: 0.7812 - lambda_1_loss: 0.1378 - dense_2_acc: 0.8781 - lambda_1_acc: 0.9701 - val_loss: 1.0523 - val_dense_2_loss: 0.9053 - val_lambda_1_loss: 0.2941 - val_dense_2_acc: 0.8101 - val_lambda_1_acc: 0.8405\n",
      "Epoch 82/100\n",
      "30000/30000 [==============================] - 5s 159us/step - loss: 0.8607 - dense_2_loss: 0.7807 - lambda_1_loss: 0.1600 - dense_2_acc: 0.8741 - lambda_1_acc: 0.9484 - val_loss: 1.0789 - val_dense_2_loss: 0.9042 - val_lambda_1_loss: 0.3496 - val_dense_2_acc: 0.7741 - val_lambda_1_acc: 0.8058\n",
      "Epoch 83/100\n",
      "30000/30000 [==============================] - 5s 159us/step - loss: 0.8278 - dense_2_loss: 0.7604 - lambda_1_loss: 0.1349 - dense_2_acc: 0.8987 - lambda_1_acc: 0.9728 - val_loss: 1.0173 - val_dense_2_loss: 0.8723 - val_lambda_1_loss: 0.2899 - val_dense_2_acc: 0.8050 - val_lambda_1_acc: 0.8597\n",
      "Epoch 84/100\n",
      "30000/30000 [==============================] - 5s 159us/step - loss: 0.8469 - dense_2_loss: 0.7640 - lambda_1_loss: 0.1659 - dense_2_acc: 0.8915 - lambda_1_acc: 0.9422 - val_loss: 1.0162 - val_dense_2_loss: 0.8575 - val_lambda_1_loss: 0.3173 - val_dense_2_acc: 0.8259 - val_lambda_1_acc: 0.8345\n",
      "Epoch 85/100\n",
      "30000/30000 [==============================] - 5s 158us/step - loss: 0.8147 - dense_2_loss: 0.7436 - lambda_1_loss: 0.1422 - dense_2_acc: 0.9105 - lambda_1_acc: 0.9642 - val_loss: 1.0383 - val_dense_2_loss: 0.8624 - val_lambda_1_loss: 0.3518 - val_dense_2_acc: 0.8058 - val_lambda_1_acc: 0.7935\n",
      "Epoch 86/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 0.7890 - dense_2_loss: 0.7269 - lambda_1_loss: 0.1244 - dense_2_acc: 0.9316 - lambda_1_acc: 0.9804 - val_loss: 0.9989 - val_dense_2_loss: 0.8377 - val_lambda_1_loss: 0.3224 - val_dense_2_acc: 0.8393 - val_lambda_1_acc: 0.8327\n",
      "Epoch 87/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 0.8020 - dense_2_loss: 0.7276 - lambda_1_loss: 0.1486 - dense_2_acc: 0.9195 - lambda_1_acc: 0.9569 - val_loss: 0.9763 - val_dense_2_loss: 0.8293 - val_lambda_1_loss: 0.2941 - val_dense_2_acc: 0.8314 - val_lambda_1_acc: 0.8581\n",
      "Epoch 88/100\n",
      "30000/30000 [==============================] - 5s 158us/step - loss: 0.7832 - dense_2_loss: 0.7140 - lambda_1_loss: 0.1385 - dense_2_acc: 0.9301 - lambda_1_acc: 0.9670 - val_loss: 0.9669 - val_dense_2_loss: 0.8077 - val_lambda_1_loss: 0.3184 - val_dense_2_acc: 0.8483 - val_lambda_1_acc: 0.8295\n",
      "Epoch 89/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 0.7627 - dense_2_loss: 0.7002 - lambda_1_loss: 0.1251 - dense_2_acc: 0.9458 - lambda_1_acc: 0.9768 - val_loss: 0.9614 - val_dense_2_loss: 0.8182 - val_lambda_1_loss: 0.2863 - val_dense_2_acc: 0.8622 - val_lambda_1_acc: 0.8614\n",
      "Epoch 90/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 0.7740 - dense_2_loss: 0.7001 - lambda_1_loss: 0.1477 - dense_2_acc: 0.9413 - lambda_1_acc: 0.9583 - val_loss: 0.9868 - val_dense_2_loss: 0.8196 - val_lambda_1_loss: 0.3344 - val_dense_2_acc: 0.8144 - val_lambda_1_acc: 0.8365\n",
      "Epoch 91/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 0.7464 - dense_2_loss: 0.6832 - lambda_1_loss: 0.1265 - dense_2_acc: 0.9550 - lambda_1_acc: 0.9745 - val_loss: 0.9660 - val_dense_2_loss: 0.8079 - val_lambda_1_loss: 0.3161 - val_dense_2_acc: 0.8492 - val_lambda_1_acc: 0.8433\n",
      "Epoch 92/100\n",
      "30000/30000 [==============================] - 5s 159us/step - loss: 0.7863 - dense_2_loss: 0.6977 - lambda_1_loss: 0.1772 - dense_2_acc: 0.9326 - lambda_1_acc: 0.9345 - val_loss: 0.9336 - val_dense_2_loss: 0.7858 - val_lambda_1_loss: 0.2957 - val_dense_2_acc: 0.8716 - val_lambda_1_acc: 0.8558\n",
      "Epoch 93/100\n",
      "30000/30000 [==============================] - 5s 159us/step - loss: 0.7339 - dense_2_loss: 0.6691 - lambda_1_loss: 0.1297 - dense_2_acc: 0.9650 - lambda_1_acc: 0.9682 - val_loss: 0.9506 - val_dense_2_loss: 0.7931 - val_lambda_1_loss: 0.3149 - val_dense_2_acc: 0.8758 - val_lambda_1_acc: 0.8382\n",
      "Epoch 94/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 0.7174 - dense_2_loss: 0.6565 - lambda_1_loss: 0.1217 - dense_2_acc: 0.9660 - lambda_1_acc: 0.9765 - val_loss: 0.9287 - val_dense_2_loss: 0.7864 - val_lambda_1_loss: 0.2847 - val_dense_2_acc: 0.8747 - val_lambda_1_acc: 0.8705\n",
      "Epoch 95/100\n",
      "30000/30000 [==============================] - 5s 159us/step - loss: 0.7277 - dense_2_loss: 0.6568 - lambda_1_loss: 0.1420 - dense_2_acc: 0.9622 - lambda_1_acc: 0.9572 - val_loss: 0.9257 - val_dense_2_loss: 0.7725 - val_lambda_1_loss: 0.3063 - val_dense_2_acc: 0.8764 - val_lambda_1_acc: 0.8483\n",
      "Epoch 96/100\n",
      "30000/30000 [==============================] - 5s 158us/step - loss: 0.6883 - dense_2_loss: 0.6348 - lambda_1_loss: 0.1070 - dense_2_acc: 0.9838 - lambda_1_acc: 0.9863 - val_loss: 0.9851 - val_dense_2_loss: 0.8024 - val_lambda_1_loss: 0.3654 - val_dense_2_acc: 0.8246 - val_lambda_1_acc: 0.8184\n",
      "Epoch 97/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 0.6855 - dense_2_loss: 0.6283 - lambda_1_loss: 0.1145 - dense_2_acc: 0.9800 - lambda_1_acc: 0.9810 - val_loss: 0.9492 - val_dense_2_loss: 0.7654 - val_lambda_1_loss: 0.3676 - val_dense_2_acc: 0.8691 - val_lambda_1_acc: 0.7989\n",
      "Epoch 98/100\n",
      "30000/30000 [==============================] - 5s 160us/step - loss: 0.6925 - dense_2_loss: 0.6267 - lambda_1_loss: 0.1317 - dense_2_acc: 0.9751 - lambda_1_acc: 0.9690 - val_loss: 0.9629 - val_dense_2_loss: 0.7829 - val_lambda_1_loss: 0.3599 - val_dense_2_acc: 0.8736 - val_lambda_1_acc: 0.8045\n",
      "Epoch 99/100\n",
      "30000/30000 [==============================] - 5s 161us/step - loss: 0.6734 - dense_2_loss: 0.6150 - lambda_1_loss: 0.1167 - dense_2_acc: 0.9848 - lambda_1_acc: 0.9779 - val_loss: 0.8904 - val_dense_2_loss: 0.7379 - val_lambda_1_loss: 0.3050 - val_dense_2_acc: 0.8855 - val_lambda_1_acc: 0.8601\n",
      "Epoch 100/100\n",
      "30000/30000 [==============================] - 5s 163us/step - loss: 0.6745 - dense_2_loss: 0.6104 - lambda_1_loss: 0.1282 - dense_2_acc: 0.9798 - lambda_1_acc: 0.9702 - val_loss: 0.8946 - val_dense_2_loss: 0.7377 - val_lambda_1_loss: 0.3138 - val_dense_2_acc: 0.8870 - val_lambda_1_acc: 0.8490\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x179e1c6a438>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dummy_matrix1=np.zeros((x_train.shape[0],1))\n",
    "dummy_matrix2=np.zeros((x_test.shape[0],1))\n",
    "\n",
    "model_center_loss.fit(x=[x_train,y_train],y=[y_train_one_hot,dummy_matrix1],\n",
    "                     batch_size=512,epochs=100,verbose=1,validation_data=([x_test,y_test],[y_test_one_hot,dummy_matrix2]),\n",
    "                     callbacks=[EarlyStopping(monitor=\"val_loss\",patience=20)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(10000, 2)\n",
      "(10, 2)\n"
     ]
    }
   ],
   "source": [
    "func=K.function([model.input],[model.get_layer('out1').output])\n",
    "test_features= func([x_test])[0]\n",
    "print(test_features.shape)\n",
    "\n",
    "test_centers=np.dot(np.transpose(y_test_one_hot),test_features)\n",
    "test_centers_count=np.sum(np.transpose(y_test_one_hot),axis=1,keepdims=True)\n",
    "test_centers/=test_centers_count\n",
    "print(test_centers.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgMAAAFkCAYAAAC9wjgoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x17c19c95eb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "%matplotlib inline\n",
    "plt.scatter(test_features[:,0],test_features[:,1],c=y_test,edgecolor=\"none\",s=5)\n",
    "plt.scatter(test_centers[:,0],test_centers[:,1],c=\"black\",marker=\"*\",edgecolor=\"none\",s=50)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda root]",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}