dyingRelu.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "https://mp.weixin.qq.com/s/XttlCNKGvGZrD7OQZOQGnw"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "x = np.random.uniform(-1, 0,  (200000, 4))\n",
    "y = np.random.randint(2, size=200000)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=0.3, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:516: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /opt/anaconda/lib/python3.7/site-packages/tensorflow/python/ops/nn_impl.py:180: add_dispatch_support.<locals>.wrapper (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.where in 2.0, which has the same broadcast rule as np.where\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:541: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:542: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:543: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:544: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:545: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:550: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n"
     ]
    }
   ],
   "source": [
    "import keras\n",
    "\n",
    "inp = keras.layers.Input(shape=(4,))\n",
    "dense_1 = keras.layers.Dense(64, activation='relu', \n",
    "                            kernel_initializer='RandomNormal', bias_initializer='zeros')(inp)\n",
    "out = keras.layers.Dense(1, activation='sigmoid')(dense_1)\n",
    "model = keras.models.Model(inp, out)\n",
    "model.compile(optimizer='Adam', loss='binary_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import datetime\n",
    "\n",
    "log_dir = \"logs/fit/\" + datetime.datetime.now().strftime(\"%Y%m%d-%H%M%S\")\n",
    "tensorboard_callback = keras.callbacks.TensorBoard(log_dir=log_dir, histogram_freq=1, write_grads=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /opt/anaconda/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py:422: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.\n",
      "\n",
      "Train on 140000 samples, validate on 60000 samples\n",
      "WARNING:tensorflow:From /opt/anaconda/lib/python3.7/site-packages/keras/callbacks/tensorboard_v1.py:198: The name tf.summary.histogram is deprecated. Please use tf.compat.v1.summary.histogram instead.\n",
      "\n",
      "WARNING:tensorflow:From /opt/anaconda/lib/python3.7/site-packages/keras/callbacks/tensorboard_v1.py:200: The name tf.summary.merge_all is deprecated. Please use tf.compat.v1.summary.merge_all instead.\n",
      "\n",
      "WARNING:tensorflow:From /opt/anaconda/lib/python3.7/site-packages/keras/callbacks/tensorboard_v1.py:203: The name tf.summary.FileWriter is deprecated. Please use tf.compat.v1.summary.FileWriter instead.\n",
      "\n",
      "Epoch 1/25\n",
      "140000/140000 [==============================] - 17s 120us/step - loss: 0.6933 - accuracy: 0.5022 - val_loss: 0.6932 - val_accuracy: 0.5019\n",
      "WARNING:tensorflow:From /opt/anaconda/lib/python3.7/site-packages/keras/callbacks/tensorboard_v1.py:343: The name tf.Summary is deprecated. Please use tf.compat.v1.Summary instead.\n",
      "\n",
      "Epoch 2/25\n",
      "140000/140000 [==============================] - 16s 113us/step - loss: 0.6932 - accuracy: 0.5015 - val_loss: 0.6932 - val_accuracy: 0.5019\n",
      "Epoch 3/25\n",
      "140000/140000 [==============================] - 16s 114us/step - loss: 0.6932 - accuracy: 0.5025 - val_loss: 0.6932 - val_accuracy: 0.4991\n",
      "Epoch 4/25\n",
      "140000/140000 [==============================] - 16s 117us/step - loss: 0.6932 - accuracy: 0.5023 - val_loss: 0.6931 - val_accuracy: 0.5028\n",
      "Epoch 5/25\n",
      "140000/140000 [==============================] - 16s 114us/step - loss: 0.6931 - accuracy: 0.5037 - val_loss: 0.6933 - val_accuracy: 0.4982\n",
      "Epoch 6/25\n",
      "140000/140000 [==============================] - 16s 113us/step - loss: 0.6932 - accuracy: 0.5041 - val_loss: 0.6932 - val_accuracy: 0.5003\n",
      "Epoch 7/25\n",
      "140000/140000 [==============================] - 16s 115us/step - loss: 0.6931 - accuracy: 0.5019 - val_loss: 0.6931 - val_accuracy: 0.5021\n",
      "Epoch 8/25\n",
      "140000/140000 [==============================] - 16s 114us/step - loss: 0.6931 - accuracy: 0.5034 - val_loss: 0.6932 - val_accuracy: 0.4996\n",
      "Epoch 9/25\n",
      "140000/140000 [==============================] - 16s 114us/step - loss: 0.6931 - accuracy: 0.5025 - val_loss: 0.6931 - val_accuracy: 0.5029\n",
      "Epoch 10/25\n",
      "140000/140000 [==============================] - 16s 114us/step - loss: 0.6931 - accuracy: 0.5025 - val_loss: 0.6932 - val_accuracy: 0.5002\n",
      "Epoch 11/25\n",
      "140000/140000 [==============================] - 16s 116us/step - loss: 0.6931 - accuracy: 0.5031 - val_loss: 0.6932 - val_accuracy: 0.5023\n",
      "Epoch 12/25\n",
      "140000/140000 [==============================] - 16s 115us/step - loss: 0.6931 - accuracy: 0.5029 - val_loss: 0.6932 - val_accuracy: 0.5006\n",
      "Epoch 13/25\n",
      "140000/140000 [==============================] - 16s 114us/step - loss: 0.6931 - accuracy: 0.5026 - val_loss: 0.6932 - val_accuracy: 0.5009\n",
      "Epoch 14/25\n",
      "140000/140000 [==============================] - 16s 114us/step - loss: 0.6931 - accuracy: 0.5033 - val_loss: 0.6931 - val_accuracy: 0.5023\n",
      "Epoch 15/25\n",
      "140000/140000 [==============================] - 16s 116us/step - loss: 0.6931 - accuracy: 0.5036 - val_loss: 0.6932 - val_accuracy: 0.5008\n",
      "Epoch 16/25\n",
      "140000/140000 [==============================] - 16s 115us/step - loss: 0.6931 - accuracy: 0.5034 - val_loss: 0.6933 - val_accuracy: 0.4984\n",
      "Epoch 17/25\n",
      "140000/140000 [==============================] - 16s 113us/step - loss: 0.6931 - accuracy: 0.5039 - val_loss: 0.6933 - val_accuracy: 0.4999\n",
      "Epoch 18/25\n",
      "140000/140000 [==============================] - 16s 113us/step - loss: 0.6931 - accuracy: 0.5032 - val_loss: 0.6932 - val_accuracy: 0.5001\n",
      "Epoch 19/25\n",
      "140000/140000 [==============================] - 16s 117us/step - loss: 0.6931 - accuracy: 0.5027 - val_loss: 0.6932 - val_accuracy: 0.4990\n",
      "Epoch 20/25\n",
      "140000/140000 [==============================] - 16s 117us/step - loss: 0.6931 - accuracy: 0.5031 - val_loss: 0.6931 - val_accuracy: 0.5033\n",
      "Epoch 21/25\n",
      "140000/140000 [==============================] - 16s 115us/step - loss: 0.6931 - accuracy: 0.5043 - val_loss: 0.6932 - val_accuracy: 0.4993\n",
      "Epoch 22/25\n",
      "140000/140000 [==============================] - 16s 114us/step - loss: 0.6931 - accuracy: 0.5042 - val_loss: 0.6933 - val_accuracy: 0.4987\n",
      "Epoch 23/25\n",
      "140000/140000 [==============================] - 16s 117us/step - loss: 0.6931 - accuracy: 0.5037 - val_loss: 0.6933 - val_accuracy: 0.4999\n",
      "Epoch 24/25\n",
      "140000/140000 [==============================] - 16s 116us/step - loss: 0.6931 - accuracy: 0.5032 - val_loss: 0.6932 - val_accuracy: 0.4985\n",
      "Epoch 25/25\n",
      "140000/140000 [==============================] - 16s 114us/step - loss: 0.6931 - accuracy: 0.5050 - val_loss: 0.6933 - val_accuracy: 0.5025\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X_train, y_train, epochs=25, verbose=True, validation_data=(X_test, y_test),\n",
    "         callbacks=[tensorboard_callback])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'val_loss': [0.6931503793398539,\n",
       "  0.6931897096633911,\n",
       "  0.6931783520698548,\n",
       "  0.6932117047309876,\n",
       "  0.6931631821632386,\n",
       "  0.6932612524350484,\n",
       "  0.693222741095225,\n",
       "  0.6932260468482971,\n",
       "  0.6931915882110595,\n",
       "  0.6931746331850688,\n",
       "  0.6932421659469604,\n",
       "  0.6931836260795593,\n",
       "  0.6932091982841492,\n",
       "  0.6931739995320638,\n",
       "  0.6931925614039103,\n",
       "  0.693193270778656,\n",
       "  0.6932424515724183,\n",
       "  0.6932115319887797,\n",
       "  0.6932076719284057,\n",
       "  0.6931829385439555,\n",
       "  0.6931840762138367,\n",
       "  0.6932127721468607,\n",
       "  0.6931901080131531,\n",
       "  0.6931847452163696,\n",
       "  0.69318467502594],\n",
       " 'val_accuracy': [0.5020333528518677,\n",
       "  0.499266654253006,\n",
       "  0.4979666769504547,\n",
       "  0.49798333644866943,\n",
       "  0.5008999705314636,\n",
       "  0.4983166754245758,\n",
       "  0.4985499978065491,\n",
       "  0.49773332476615906,\n",
       "  0.4969833195209503,\n",
       "  0.4988499879837036,\n",
       "  0.49763333797454834,\n",
       "  0.4999333322048187,\n",
       "  0.49900001287460327,\n",
       "  0.5013999938964844,\n",
       "  0.49736666679382324,\n",
       "  0.5002833604812622,\n",
       "  0.4983333349227905,\n",
       "  0.5028333067893982,\n",
       "  0.4976166784763336,\n",
       "  0.5023000240325928,\n",
       "  0.5029333233833313,\n",
       "  0.5016166567802429,\n",
       "  0.49728333950042725,\n",
       "  0.5004333257675171,\n",
       "  0.49843332171440125],\n",
       " 'loss': [0.6931691573143005,\n",
       "  0.6931639513015747,\n",
       "  0.6931584427697318,\n",
       "  0.6931707314491272,\n",
       "  0.6931588725634984,\n",
       "  0.6931477743966239,\n",
       "  0.6931581681387765,\n",
       "  0.693159971673148,\n",
       "  0.6931557769639152,\n",
       "  0.6931487321853638,\n",
       "  0.6931553644725255,\n",
       "  0.6931544614928109,\n",
       "  0.6931567063331604,\n",
       "  0.6931505510193962,\n",
       "  0.6931411285945347,\n",
       "  0.6931472991670881,\n",
       "  0.6931481104442051,\n",
       "  0.6931509510449001,\n",
       "  0.6931525010653905,\n",
       "  0.6931625687735421,\n",
       "  0.6931359748022897,\n",
       "  0.6931455115863255,\n",
       "  0.6931569870812553,\n",
       "  0.6931459781374251,\n",
       "  0.693163267149244],\n",
       " 'accuracy': [0.5000643,\n",
       "  0.50125,\n",
       "  0.50312144,\n",
       "  0.49944285,\n",
       "  0.50082856,\n",
       "  0.50175714,\n",
       "  0.50144285,\n",
       "  0.5014643,\n",
       "  0.5025143,\n",
       "  0.50184286,\n",
       "  0.5019929,\n",
       "  0.5023357,\n",
       "  0.50189286,\n",
       "  0.50203574,\n",
       "  0.50222856,\n",
       "  0.50144285,\n",
       "  0.50435716,\n",
       "  0.50116426,\n",
       "  0.50058573,\n",
       "  0.50042856,\n",
       "  0.5025143,\n",
       "  0.5021286,\n",
       "  0.5017357,\n",
       "  0.50264287,\n",
       "  0.5004929]}"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "history.history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Epochs')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.ylabel('Val loss')\n",
    "plt.xlabel('Epochs')\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Epochs')"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['loss'])\n",
    "plt.ylabel('Training loss')\n",
    "plt.xlabel('Epochs')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 添加层"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:516: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /opt/anaconda/lib/python3.7/site-packages/tensorflow/python/ops/nn_impl.py:180: add_dispatch_support.<locals>.wrapper (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.where in 2.0, which has the same broadcast rule as np.where\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:541: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:542: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:543: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:544: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:545: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:550: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n"
     ]
    }
   ],
   "source": [
    "import keras\n",
    "\n",
    "inp = keras.layers.Input(shape=(4,))\n",
    "dense_1 = keras.layers.Dense(64, activation='relu', \n",
    "                            kernel_initializer='RandomNormal', bias_initializer='zeros')(inp)\n",
    "dense_2 = keras.layers.Dense(64, activation='relu', \n",
    "                            kernel_initializer='RandomNormal', bias_initializer='zeros')(dense_1)\n",
    "dense_3 = keras.layers.Dense(64, activation='relu', \n",
    "                            kernel_initializer='RandomNormal', bias_initializer='zeros')(dense_2)\n",
    "out = keras.layers.Dense(1, activation='sigmoid')(dense_3)\n",
    "model = keras.models.Model(inp, out)\n",
    "model.compile(optimizer='Adam', loss='binary_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import datetime\n",
    "\n",
    "log_dir = \"logs/fit/\" + datetime.datetime.now().strftime(\"%Y%m%d-%H%M%S\")\n",
    "tensorboard_callback = keras.callbacks.TensorBoard(log_dir=log_dir, histogram_freq=1, write_grads=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 140000 samples, validate on 60000 samples\n",
      "Epoch 1/25\n",
      "140000/140000 [==============================] - 21s 152us/step - loss: 0.6932 - accuracy: 0.5009 - val_loss: 0.6931 - val_accuracy: 0.5013\n",
      "Epoch 2/25\n",
      "140000/140000 [==============================] - 21s 152us/step - loss: 0.6932 - accuracy: 0.5004 - val_loss: 0.6932 - val_accuracy: 0.5013\n",
      "Epoch 3/25\n",
      "140000/140000 [==============================] - 21s 151us/step - loss: 0.6932 - accuracy: 0.5021 - val_loss: 0.6932 - val_accuracy: 0.5013\n",
      "Epoch 4/25\n",
      "140000/140000 [==============================] - 21s 150us/step - loss: 0.6932 - accuracy: 0.4991 - val_loss: 0.6932 - val_accuracy: 0.5013\n",
      "Epoch 5/25\n",
      "140000/140000 [==============================] - 20s 145us/step - loss: 0.6932 - accuracy: 0.4983 - val_loss: 0.6932 - val_accuracy: 0.4987\n",
      "Epoch 6/25\n",
      "140000/140000 [==============================] - 21s 148us/step - loss: 0.6932 - accuracy: 0.5001 - val_loss: 0.6932 - val_accuracy: 0.4987\n",
      "Epoch 7/25\n",
      "140000/140000 [==============================] - 21s 151us/step - loss: 0.6932 - accuracy: 0.5016 - val_loss: 0.6932 - val_accuracy: 0.4987\n",
      "Epoch 8/25\n",
      "140000/140000 [==============================] - 21s 149us/step - loss: 0.6932 - accuracy: 0.4999 - val_loss: 0.6932 - val_accuracy: 0.5013\n",
      "Epoch 9/25\n",
      "140000/140000 [==============================] - 21s 150us/step - loss: 0.6932 - accuracy: 0.4986 - val_loss: 0.6932 - val_accuracy: 0.4987\n",
      "Epoch 10/25\n",
      "140000/140000 [==============================] - 21s 151us/step - loss: 0.6931 - accuracy: 0.5009 - val_loss: 0.6932 - val_accuracy: 0.5013\n",
      "Epoch 11/25\n",
      "140000/140000 [==============================] - 21s 148us/step - loss: 0.6932 - accuracy: 0.4986 - val_loss: 0.6931 - val_accuracy: 0.5013\n",
      "Epoch 12/25\n",
      "140000/140000 [==============================] - 21s 149us/step - loss: 0.6932 - accuracy: 0.5018 - val_loss: 0.6931 - val_accuracy: 0.5013\n",
      "Epoch 13/25\n",
      "140000/140000 [==============================] - 21s 149us/step - loss: 0.6932 - accuracy: 0.4997 - val_loss: 0.6932 - val_accuracy: 0.5013\n",
      "Epoch 14/25\n",
      "140000/140000 [==============================] - 21s 149us/step - loss: 0.6932 - accuracy: 0.5004 - val_loss: 0.6931 - val_accuracy: 0.5013\n",
      "Epoch 15/25\n",
      "140000/140000 [==============================] - 21s 149us/step - loss: 0.6932 - accuracy: 0.4968 - val_loss: 0.6932 - val_accuracy: 0.4987\n",
      "Epoch 16/25\n",
      "140000/140000 [==============================] - 21s 149us/step - loss: 0.6932 - accuracy: 0.5006 - val_loss: 0.6932 - val_accuracy: 0.4987\n",
      "Epoch 17/25\n",
      "140000/140000 [==============================] - 21s 152us/step - loss: 0.6932 - accuracy: 0.5001 - val_loss: 0.6932 - val_accuracy: 0.4987\n",
      "Epoch 18/25\n",
      "140000/140000 [==============================] - 21s 149us/step - loss: 0.6932 - accuracy: 0.5010 - val_loss: 0.6932 - val_accuracy: 0.4987\n",
      "Epoch 19/25\n",
      "140000/140000 [==============================] - 21s 149us/step - loss: 0.6932 - accuracy: 0.5000 - val_loss: 0.6931 - val_accuracy: 0.5013\n",
      "Epoch 20/25\n",
      "140000/140000 [==============================] - 21s 150us/step - loss: 0.6932 - accuracy: 0.5000 - val_loss: 0.6932 - val_accuracy: 0.4987\n",
      "Epoch 21/25\n",
      "140000/140000 [==============================] - 21s 152us/step - loss: 0.6932 - accuracy: 0.5005 - val_loss: 0.6932 - val_accuracy: 0.4987\n",
      "Epoch 22/25\n",
      "140000/140000 [==============================] - 21s 147us/step - loss: 0.6932 - accuracy: 0.5014 - val_loss: 0.6932 - val_accuracy: 0.4987\n",
      "Epoch 23/25\n",
      "140000/140000 [==============================] - 21s 150us/step - loss: 0.6932 - accuracy: 0.4992 - val_loss: 0.6931 - val_accuracy: 0.4987\n",
      "Epoch 24/25\n",
      "140000/140000 [==============================] - 22s 154us/step - loss: 0.6932 - accuracy: 0.5001 - val_loss: 0.6932 - val_accuracy: 0.4987\n",
      "Epoch 25/25\n",
      "140000/140000 [==============================] - 21s 149us/step - loss: 0.6932 - accuracy: 0.4999 - val_loss: 0.6932 - val_accuracy: 0.4987\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X_train, y_train, epochs=25, verbose=True, validation_data=(X_test, y_test),\n",
    "         callbacks=[tensorboard_callback])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 改变激活函数"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:516: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /opt/anaconda/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py:3313: add_dispatch_support.<locals>.wrapper (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.where in 2.0, which has the same broadcast rule as np.where\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:541: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:542: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:543: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:544: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:545: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n",
      "/opt/anaconda/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:550: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n"
     ]
    }
   ],
   "source": [
    "import keras\n",
    "\n",
    "inp = keras.layers.Input(shape=(4,))\n",
    "dense_1 = keras.layers.Dense(64, activation='selu', kernel_initializer='lecun_normal', \n",
    "                  bias_initializer='lecun_normal')(inp)\n",
    "alp_drop = keras.layers.AlphaDropout(0.25)(dense_1)\n",
    "out = keras.layers.Dense(1, activation='sigmoid')(alp_drop)\n",
    "model = keras.models.Model(inp, out)\n",
    "model.compile(optimizer='Adam', loss='binary_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /opt/anaconda/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py:422: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.\n",
      "\n",
      "Train on 140000 samples, validate on 60000 samples\n",
      "WARNING:tensorflow:From /opt/anaconda/lib/python3.7/site-packages/keras/callbacks/tensorboard_v1.py:198: The name tf.summary.histogram is deprecated. Please use tf.compat.v1.summary.histogram instead.\n",
      "\n",
      "WARNING:tensorflow:From /opt/anaconda/lib/python3.7/site-packages/keras/callbacks/tensorboard_v1.py:200: The name tf.summary.merge_all is deprecated. Please use tf.compat.v1.summary.merge_all instead.\n",
      "\n",
      "WARNING:tensorflow:From /opt/anaconda/lib/python3.7/site-packages/keras/callbacks/tensorboard_v1.py:203: The name tf.summary.FileWriter is deprecated. Please use tf.compat.v1.summary.FileWriter instead.\n",
      "\n",
      "Epoch 1/25\n",
      "140000/140000 [==============================] - 19s 139us/step - loss: 0.7016 - accuracy: 0.5006 - val_loss: 0.6936 - val_accuracy: 0.5022\n",
      "WARNING:tensorflow:From /opt/anaconda/lib/python3.7/site-packages/keras/callbacks/tensorboard_v1.py:343: The name tf.Summary is deprecated. Please use tf.compat.v1.Summary instead.\n",
      "\n",
      "Epoch 2/25\n",
      "140000/140000 [==============================] - 19s 136us/step - loss: 0.6943 - accuracy: 0.5022 - val_loss: 0.6945 - val_accuracy: 0.5016\n",
      "Epoch 3/25\n",
      "140000/140000 [==============================] - 19s 135us/step - loss: 0.6942 - accuracy: 0.4992 - val_loss: 0.6939 - val_accuracy: 0.4976\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-5-be0d0c3ec520>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mtensorboard_callback\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mkeras\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcallbacks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTensorBoard\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlog_dir\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mlog_dir\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhistogram_freq\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwrite_grads\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m history = model.fit(X_train, y_train, epochs=25, verbose=1, validation_data=(X_test, y_test),\n\u001b[0;32m----> 6\u001b[0;31m          callbacks=[tensorboard_callback])\n\u001b[0m",
      "\u001b[0;32m/opt/anaconda/lib/python3.7/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq, max_queue_size, workers, use_multiprocessing, **kwargs)\u001b[0m\n\u001b[1;32m   1237\u001b[0m                                         \u001b[0msteps_per_epoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msteps_per_epoch\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1238\u001b[0m                                         \u001b[0mvalidation_steps\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mvalidation_steps\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1239\u001b[0;31m                                         validation_freq=validation_freq)\n\u001b[0m\u001b[1;32m   1240\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1241\u001b[0m     def evaluate(self,\n",
      "\u001b[0;32m/opt/anaconda/lib/python3.7/site-packages/keras/engine/training_arrays.py\u001b[0m in \u001b[0;36mfit_loop\u001b[0;34m(model, fit_function, fit_inputs, out_labels, batch_size, epochs, verbose, callbacks, val_function, val_inputs, shuffle, initial_epoch, steps_per_epoch, validation_steps, validation_freq)\u001b[0m\n\u001b[1;32m    214\u001b[0m                         \u001b[0mepoch_logs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'val_'\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0ml\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mo\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    215\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 216\u001b[0;31m         \u001b[0mcallbacks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_epoch_end\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mepoch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mepoch_logs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    217\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mcallbacks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstop_training\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    218\u001b[0m             \u001b[0;32mbreak\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda/lib/python3.7/site-packages/keras/callbacks/callbacks.py\u001b[0m in \u001b[0;36mon_epoch_end\u001b[0;34m(self, epoch, logs)\u001b[0m\n\u001b[1;32m    150\u001b[0m         \u001b[0mlogs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlogs\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0;34m{\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    151\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mcallback\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcallbacks\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 152\u001b[0;31m             \u001b[0mcallback\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_epoch_end\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mepoch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlogs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    153\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    154\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mon_train_batch_begin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlogs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda/lib/python3.7/site-packages/keras/callbacks/tensorboard_v1.py\u001b[0m in \u001b[0;36mon_epoch_end\u001b[0;34m(self, epoch, logs)\u001b[0m\n\u001b[1;32m    287\u001b[0m                     \u001b[0;32massert\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch_val\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtensors\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    288\u001b[0m                     \u001b[0mfeed_dict\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mzip\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtensors\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_val\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 289\u001b[0;31m                     \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msess\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmerged\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfeed_dict\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    290\u001b[0m                     \u001b[0msummary_str\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mresult\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    291\u001b[0m                     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwriter\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd_summary\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msummary_str\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mepoch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36mrun\u001b[0;34m(self, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m    948\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    949\u001b[0m       result = self._run(None, fetches, feed_dict, options_ptr,\n\u001b[0;32m--> 950\u001b[0;31m                          run_metadata_ptr)\n\u001b[0m\u001b[1;32m    951\u001b[0m       \u001b[0;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    952\u001b[0m         \u001b[0mproto_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_session\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTF_GetBuffer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrun_metadata_ptr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run\u001b[0;34m(self, handle, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m   1171\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mfinal_fetches\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mfinal_targets\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mhandle\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mfeed_dict_tensor\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1172\u001b[0m       results = self._do_run(handle, final_targets, final_fetches,\n\u001b[0;32m-> 1173\u001b[0;31m                              feed_dict_tensor, options, run_metadata)\n\u001b[0m\u001b[1;32m   1174\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1175\u001b[0m       \u001b[0mresults\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_run\u001b[0;34m(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m   1348\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mhandle\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1349\u001b[0m       return self._do_call(_run_fn, feeds, fetches, targets, options,\n\u001b[0;32m-> 1350\u001b[0;31m                            run_metadata)\n\u001b[0m\u001b[1;32m   1351\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1352\u001b[0m       \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_prun_fn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeeds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetches\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_call\u001b[0;34m(self, fn, *args)\u001b[0m\n\u001b[1;32m   1354\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0m_do_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1355\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1356\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1357\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0merrors\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mOpError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1358\u001b[0m       \u001b[0mmessage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcompat\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mas_text\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmessage\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run_fn\u001b[0;34m(feed_dict, fetch_list, target_list, options, run_metadata)\u001b[0m\n\u001b[1;32m   1339\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_extend_graph\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1340\u001b[0m       return self._call_tf_sessionrun(\n\u001b[0;32m-> 1341\u001b[0;31m           options, feed_dict, fetch_list, target_list, run_metadata)\n\u001b[0m\u001b[1;32m   1342\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1343\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_prun_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda/lib/python3.7/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_call_tf_sessionrun\u001b[0;34m(self, options, feed_dict, fetch_list, target_list, run_metadata)\u001b[0m\n\u001b[1;32m   1427\u001b[0m     return tf_session.TF_SessionRun_wrapper(\n\u001b[1;32m   1428\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_session\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moptions\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget_list\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1429\u001b[0;31m         run_metadata)\n\u001b[0m\u001b[1;32m   1430\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1431\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0m_call_tf_sessionprun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "import datetime\n",
    "\n",
    "log_dir = \"logs/fit/\" + datetime.datetime.now().strftime(\"%Y%m%d-%H%M%S\")\n",
    "tensorboard_callback = keras.callbacks.TensorBoard(log_dir=log_dir, histogram_freq=1, write_grads=True)\n",
    "history = model.fit(X_train, y_train, epochs=25, verbose=1, validation_data=(X_test, y_test),\n",
    "         callbacks=[tensorboard_callback])"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}