002-keras_function.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# keras 函数api"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# !pip install pydot\n",
    "#!sudo apt-get install graphvizf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from __future__ import absolute_import, division, print_function\n",
    "import tensorflow as tf\n",
    "import tensorflow.keras as keras\n",
    "import tensorflow.keras.layers as layers\n",
    "tf.keras.backend.clear_session()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1构建简单的网络\n",
    "### 1.1创建网络"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"mnist model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "img (InputLayer)             [(None, 784)]             0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 32)                25120     \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 32)                1056      \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 10)                330       \n",
      "=================================================================\n",
      "Total params: 26,506\n",
      "Trainable params: 26,506\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "inputs = tf.keras.Input(shape=(784,), name='img')\n",
    "h1 = layers.Dense(32, activation='relu')(inputs)\n",
    "h2 = layers.Dense(32, activation='relu')(h1)\n",
    "outputs = layers.Dense(10, activation='softmax')(h2)\n",
    "model = tf.keras.Model(inputs=inputs, outputs=outputs, name='mnist model')\n",
    "\n",
    "model.summary()\n",
    "keras.utils.plot_model(model, 'mnist_model.png')\n",
    "keras.utils.plot_model(model, 'model_info.png', show_shapes=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.2训练、验证及测试"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 48000 samples, validate on 12000 samples\n",
      "Epoch 1/5\n",
      "48000/48000 [==============================] - 1s 25us/sample - loss: 0.4348 - accuracy: 0.8789 - val_loss: 0.2452 - val_accuracy: 0.9301\n",
      "Epoch 2/5\n",
      "48000/48000 [==============================] - 1s 20us/sample - loss: 0.2164 - accuracy: 0.9371 - val_loss: 0.1792 - val_accuracy: 0.9487\n",
      "Epoch 3/5\n",
      "48000/48000 [==============================] - 1s 20us/sample - loss: 0.1706 - accuracy: 0.9501 - val_loss: 0.1632 - val_accuracy: 0.9532\n",
      "Epoch 4/5\n",
      "48000/48000 [==============================] - 1s 20us/sample - loss: 0.1438 - accuracy: 0.9572 - val_loss: 0.1585 - val_accuracy: 0.9553\n",
      "Epoch 5/5\n",
      "48000/48000 [==============================] - 1s 20us/sample - loss: 0.1261 - accuracy: 0.9627 - val_loss: 0.1381 - val_accuracy: 0.9603\n",
      "test loss: 0.14040755774863065\n",
      "test acc: 0.9605\n"
     ]
    }
   ],
   "source": [
    "(x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data()\n",
    "x_train = x_train.reshape(60000, 784).astype('float32') /255\n",
    "x_test = x_test.reshape(10000, 784).astype('float32') /255\n",
    "model.compile(optimizer=keras.optimizers.RMSprop(),\n",
    "             loss='sparse_categorical_crossentropy', # 直接填api，后面会报错\n",
    "             metrics=['accuracy'])\n",
    "history = model.fit(x_train, y_train, batch_size=64, epochs=5, validation_split=0.2)\n",
    "test_scores = model.evaluate(x_test, y_test, verbose=0)\n",
    "print('test loss:', test_scores[0])\n",
    "print('test acc:', test_scores[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.3模型保持和序列化"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('model_save.h5')\n",
    "del model\n",
    "model = keras.models.load_model('model_save.h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2 使用共享网络创建多个模型\n",
    "在函数API中，通过在图层图中指定其输入和输出来创建模型。 这意味着可以使用单个图层图来生成多个模型。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"encoder\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "img (InputLayer)             [(None, 28, 28, 1)]       0         \n",
      "_________________________________________________________________\n",
      "conv2d_8 (Conv2D)            (None, 26, 26, 16)        160       \n",
      "_________________________________________________________________\n",
      "conv2d_9 (Conv2D)            (None, 24, 24, 32)        4640      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 8, 8, 32)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_10 (Conv2D)           (None, 6, 6, 32)          9248      \n",
      "_________________________________________________________________\n",
      "conv2d_11 (Conv2D)           (None, 4, 4, 16)          4624      \n",
      "_________________________________________________________________\n",
      "global_max_pooling2d_1 (Glob (None, 16)                0         \n",
      "=================================================================\n",
      "Total params: 18,672\n",
      "Trainable params: 18,672\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "Model: \"autoencoder\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "img (InputLayer)             [(None, 28, 28, 1)]       0         \n",
      "_________________________________________________________________\n",
      "conv2d_8 (Conv2D)            (None, 26, 26, 16)        160       \n",
      "_________________________________________________________________\n",
      "conv2d_9 (Conv2D)            (None, 24, 24, 32)        4640      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 8, 8, 32)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_10 (Conv2D)           (None, 6, 6, 32)          9248      \n",
      "_________________________________________________________________\n",
      "conv2d_11 (Conv2D)           (None, 4, 4, 16)          4624      \n",
      "_________________________________________________________________\n",
      "global_max_pooling2d_1 (Glob (None, 16)                0         \n",
      "_________________________________________________________________\n",
      "reshape (Reshape)            (None, 4, 4, 1)           0         \n",
      "_________________________________________________________________\n",
      "conv2d_transpose (Conv2DTran (None, 6, 6, 16)          160       \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_1 (Conv2DTr (None, 8, 8, 32)          4640      \n",
      "_________________________________________________________________\n",
      "up_sampling2d (UpSampling2D) (None, 24, 24, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_2 (Conv2DTr (None, 26, 26, 16)        4624      \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_3 (Conv2DTr (None, 28, 28, 1)         145       \n",
      "=================================================================\n",
      "Total params: 28,241\n",
      "Trainable params: 28,241\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# 编码器网络和自编码器网络\n",
    "\n",
    "encode_input = keras.Input(shape=(28,28,1), name='img')\n",
    "h1 = layers.Conv2D(16, 3, activation='relu')(encode_input)\n",
    "h1 = layers.Conv2D(32, 3, activation='relu')(h1)\n",
    "h1 = layers.MaxPool2D(3)(h1)\n",
    "h1 = layers.Conv2D(32, 3, activation='relu')(h1)\n",
    "h1 = layers.Conv2D(16, 3, activation='relu')(h1)\n",
    "encode_output = layers.GlobalMaxPool2D()(h1)\n",
    "\n",
    "encode_model = keras.Model(inputs=encode_input, outputs=encode_output, name='encoder')\n",
    "encode_model.summary()\n",
    "\n",
    "h2 = layers.Reshape((4, 4, 1))(encode_output)\n",
    "h2 = layers.Conv2DTranspose(16, 3, activation='relu')(h2)\n",
    "h2 = layers.Conv2DTranspose(32, 3, activation='relu')(h2)\n",
    "h2 = layers.UpSampling2D(3)(h2)\n",
    "h2 = layers.Conv2DTranspose(16, 3, activation='relu')(h2)\n",
    "decode_output = layers.Conv2DTranspose(1, 3, activation='relu')(h2)\n",
    "\n",
    "autoencoder = keras.Model(inputs=encode_input, outputs=decode_output, name='autoencoder')\n",
    "autoencoder.summary()\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "可以把整个模型，当作一层网络使用"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"encoder\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "src_img (InputLayer)         [(None, 28, 28, 1)]       0         \n",
      "_________________________________________________________________\n",
      "conv2d_16 (Conv2D)           (None, 26, 26, 16)        160       \n",
      "_________________________________________________________________\n",
      "conv2d_17 (Conv2D)           (None, 24, 24, 32)        4640      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_4 (MaxPooling2 (None, 8, 8, 32)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_18 (Conv2D)           (None, 6, 6, 32)          9248      \n",
      "_________________________________________________________________\n",
      "conv2d_19 (Conv2D)           (None, 4, 4, 16)          4624      \n",
      "_________________________________________________________________\n",
      "global_max_pooling2d_3 (Glob (None, 16)                0         \n",
      "=================================================================\n",
      "Total params: 18,672\n",
      "Trainable params: 18,672\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "Model: \"decoder\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "encoded_img (InputLayer)     [(None, 16)]              0         \n",
      "_________________________________________________________________\n",
      "reshape_2 (Reshape)          (None, 4, 4, 1)           0         \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_8 (Conv2DTr (None, 6, 6, 16)          160       \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_9 (Conv2DTr (None, 8, 8, 32)          4640      \n",
      "_________________________________________________________________\n",
      "up_sampling2d_2 (UpSampling2 (None, 24, 24, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_10 (Conv2DT (None, 26, 26, 16)        4624      \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_11 (Conv2DT (None, 28, 28, 1)         145       \n",
      "=================================================================\n",
      "Total params: 9,569\n",
      "Trainable params: 9,569\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "Model: \"autoencoder\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "img (InputLayer)             [(None, 28, 28, 1)]       0         \n",
      "_________________________________________________________________\n",
      "encoder (Model)              (None, 16)                18672     \n",
      "_________________________________________________________________\n",
      "decoder (Model)              (None, 28, 28, 1)         9569      \n",
      "=================================================================\n",
      "Total params: 28,241\n",
      "Trainable params: 28,241\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "encode_input = keras.Input(shape=(28,28,1), name='src_img')\n",
    "h1 = layers.Conv2D(16, 3, activation='relu')(encode_input)\n",
    "h1 = layers.Conv2D(32, 3, activation='relu')(h1)\n",
    "h1 = layers.MaxPool2D(3)(h1)\n",
    "h1 = layers.Conv2D(32, 3, activation='relu')(h1)\n",
    "h1 = layers.Conv2D(16, 3, activation='relu')(h1)\n",
    "encode_output = layers.GlobalMaxPool2D()(h1)\n",
    "\n",
    "encode_model = keras.Model(inputs=encode_input, outputs=encode_output, name='encoder')\n",
    "encode_model.summary()\n",
    "\n",
    "decode_input = keras.Input(shape=(16,), name='encoded_img')\n",
    "h2 = layers.Reshape((4, 4, 1))(decode_input)\n",
    "h2 = layers.Conv2DTranspose(16, 3, activation='relu')(h2)\n",
    "h2 = layers.Conv2DTranspose(32, 3, activation='relu')(h2)\n",
    "h2 = layers.UpSampling2D(3)(h2)\n",
    "h2 = layers.Conv2DTranspose(16, 3, activation='relu')(h2)\n",
    "decode_output = layers.Conv2DTranspose(1, 3, activation='relu')(h2)\n",
    "decode_model = keras.Model(inputs=decode_input, outputs=decode_output, name='decoder')\n",
    "decode_model.summary()\n",
    "\n",
    "autoencoder_input = keras.Input(shape=(28,28,1), name='img')\n",
    "h3 = encode_model(autoencoder_input)\n",
    "autoencoder_output = decode_model(h3)\n",
    "autoencoder = keras.Model(inputs=autoencoder_input, outputs=autoencoder_output,\n",
    "                          name='autoencoder')\n",
    "autoencoder.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.复杂网络结构构建\n",
    "### 3.1多输入与多输出网络"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_1\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "title (InputLayer)              [(None, None)]       0                                            \n",
      "__________________________________________________________________________________________________\n",
      "body (InputLayer)               [(None, None)]       0                                            \n",
      "__________________________________________________________________________________________________\n",
      "embedding_7 (Embedding)         (None, None, 64)     128000      title[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "embedding_6 (Embedding)         (None, None, 64)     128000      body[0][0]                       \n",
      "__________________________________________________________________________________________________\n",
      "unified_lstm_7 (UnifiedLSTM)    (None, 128)          98816       embedding_7[0][0]                \n",
      "__________________________________________________________________________________________________\n",
      "unified_lstm_6 (UnifiedLSTM)    (None, 32)           12416       embedding_6[0][0]                \n",
      "__________________________________________________________________________________________________\n",
      "tag (InputLayer)                [(None, 12)]         0                                            \n",
      "__________________________________________________________________________________________________\n",
      "concatenate_2 (Concatenate)     (None, 172)          0           unified_lstm_7[0][0]             \n",
      "                                                                 unified_lstm_6[0][0]             \n",
      "                                                                 tag[0][0]                        \n",
      "__________________________________________________________________________________________________\n",
      "priority (Dense)                (None, 1)            173         concatenate_2[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "department (Dense)              (None, 4)            692         concatenate_2[0][0]              \n",
      "==================================================================================================\n",
      "Total params: 368,097\n",
      "Trainable params: 368,097\n",
      "Non-trainable params: 0\n",
      "__________________________________________________________________________________________________\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 构建一个根据文档内容、标签和标题，预测文档优先级和执行部门的网络\n",
    "# 超参\n",
    "num_words = 2000\n",
    "num_tags = 12\n",
    "num_departments = 4\n",
    "\n",
    "# 输入\n",
    "body_input = keras.Input(shape=(None,), name='body')\n",
    "title_input = keras.Input(shape=(None,), name='title')\n",
    "tag_input = keras.Input(shape=(num_tags,), name='tag')\n",
    "\n",
    "# 嵌入层\n",
    "body_feat = layers.Embedding(num_words, 64)(body_input)\n",
    "title_feat = layers.Embedding(num_words, 64)(title_input)\n",
    "\n",
    "# 特征提取层\n",
    "body_feat = layers.LSTM(32)(body_feat)\n",
    "title_feat = layers.LSTM(128)(title_feat)\n",
    "features = layers.concatenate([title_feat,body_feat, tag_input])\n",
    "\n",
    "# 分类层\n",
    "priority_pred = layers.Dense(1, activation='sigmoid', name='priority')(features)\n",
    "department_pred = layers.Dense(num_departments, activation='softmax', name='department')(features)\n",
    "\n",
    "\n",
    "# 构建模型\n",
    "model = keras.Model(inputs=[body_input, title_input, tag_input],\n",
    "                    outputs=[priority_pred, department_pred])\n",
    "model.summary()\n",
    "keras.utils.plot_model(model, 'multi_model.png', show_shapes=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "1280/1280 [==============================] - 7s 6ms/sample - loss: 1.2544 - priority_loss: 0.6996 - department_loss: 2.7743\n",
      "Epoch 2/5\n",
      "1280/1280 [==============================] - 7s 5ms/sample - loss: 1.2175 - priority_loss: 0.6682 - department_loss: 2.7468\n",
      "Epoch 3/5\n",
      "1280/1280 [==============================] - 7s 6ms/sample - loss: 1.1374 - priority_loss: 0.5954 - department_loss: 2.7098\n",
      "Epoch 4/5\n",
      "1280/1280 [==============================] - 7s 6ms/sample - loss: 1.0794 - priority_loss: 0.5482 - department_loss: 2.6561\n",
      "Epoch 5/5\n",
      "1280/1280 [==============================] - 7s 5ms/sample - loss: 1.0388 - priority_loss: 0.5233 - department_loss: 2.5774\n"
     ]
    }
   ],
   "source": [
    "model.compile(optimizer=keras.optimizers.RMSprop(1e-3),\n",
    "             loss={'priority': 'binary_crossentropy',\n",
    "                  'department': 'categorical_crossentropy'},\n",
    "             loss_weights=[1., 0.2])\n",
    "\n",
    "import numpy as np\n",
    "# 载入输入数据\n",
    "title_data = np.random.randint(num_words, size=(1280, 10))\n",
    "body_data = np.random.randint(num_words, size=(1280, 100))\n",
    "tag_data = np.random.randint(2, size=(1280, num_tags)).astype('float32')\n",
    "# 标签\n",
    "priority_label = np.random.random(size=(1280, 1))\n",
    "department_label = np.random.randint(2, size=(1280, num_departments))\n",
    "# 训练\n",
    "history = model.fit(\n",
    "    {'title': title_data, 'body':body_data, 'tag':tag_data},\n",
    "    {'priority':priority_label, 'department':department_label},\n",
    "    batch_size=32,\n",
    "    epochs=5\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.2小型残差网络"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"small resnet\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "img (InputLayer)                [(None, 32, 32, 3)]  0                                            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_28 (Conv2D)              (None, 30, 30, 32)   896         img[0][0]                        \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_29 (Conv2D)              (None, 28, 28, 64)   18496       conv2d_28[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_7 (MaxPooling2D)  (None, 9, 9, 64)     0           conv2d_29[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_30 (Conv2D)              (None, 9, 9, 64)     36928       max_pooling2d_7[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_31 (Conv2D)              (None, 9, 9, 64)     36928       conv2d_30[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "add_2 (Add)                     (None, 9, 9, 64)     0           conv2d_31[0][0]                  \n",
      "                                                                 max_pooling2d_7[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_32 (Conv2D)              (None, 9, 9, 64)     36928       add_2[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_33 (Conv2D)              (None, 9, 9, 64)     36928       conv2d_32[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "add_3 (Add)                     (None, 9, 9, 64)     0           conv2d_33[0][0]                  \n",
      "                                                                 add_2[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_34 (Conv2D)              (None, 7, 7, 64)     36928       add_3[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "global_max_pooling2d_4 (GlobalM (None, 64)           0           conv2d_34[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "dense_7 (Dense)                 (None, 256)          16640       global_max_pooling2d_4[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "dropout (Dropout)               (None, 256)          0           dense_7[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "dense_8 (Dense)                 (None, 10)           2570        dropout[0][0]                    \n",
      "==================================================================================================\n",
      "Total params: 223,242\n",
      "Trainable params: 223,242\n",
      "Non-trainable params: 0\n",
      "__________________________________________________________________________________________________\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inputs = keras.Input(shape=(32,32,3), name='img')\n",
    "h1 = layers.Conv2D(32, 3, activation='relu')(inputs)\n",
    "h1 = layers.Conv2D(64, 3, activation='relu')(h1)\n",
    "block1_out = layers.MaxPooling2D(3)(h1)\n",
    "\n",
    "h2 = layers.Conv2D(64, 3, activation='relu', padding='same')(block1_out)\n",
    "h2 = layers.Conv2D(64, 3, activation='relu', padding='same')(h2)\n",
    "block2_out = layers.add([h2, block1_out])\n",
    "\n",
    "h3 = layers.Conv2D(64, 3, activation='relu', padding='same')(block2_out)\n",
    "h3 = layers.Conv2D(64, 3, activation='relu', padding='same')(h3)\n",
    "block3_out = layers.add([h3, block2_out])\n",
    "\n",
    "h4 = layers.Conv2D(64, 3, activation='relu')(block3_out)\n",
    "h4 = layers.GlobalMaxPool2D()(h4)\n",
    "h4 = layers.Dense(256, activation='relu')(h4)\n",
    "h4 = layers.Dropout(0.5)(h4)\n",
    "outputs = layers.Dense(10, activation='softmax')(h4)\n",
    "\n",
    "model = keras.Model(inputs, outputs, name='small resnet')\n",
    "model.summary()\n",
    "keras.utils.plot_model(model, 'small_resnet_model.png', show_shapes=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 40000 samples, validate on 10000 samples\n",
      "40000/40000 [==============================] - 85s 2ms/sample - loss: 1.4249 - acc: 0.4866 - val_loss: 1.2088 - val_acc: 0.5632\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x7fb4d2cdacf8>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(x_train, y_train), (x_test, y_test) = keras.datasets.cifar10.load_data()\n",
    "x_train = x_train.astype('float32') / 255\n",
    "x_test = y_train.astype('float32') / 255\n",
    "y_train = keras.utils.to_categorical(y_train, 10)\n",
    "y_test = keras.utils.to_categorical(y_test, 10)\n",
    "\n",
    "model.compile(optimizer=keras.optimizers.RMSprop(1e-3),\n",
    "             loss='categorical_crossentropy',\n",
    "             metrics=['acc'])\n",
    "model.fit(x_train, y_train,\n",
    "         batch_size=64,\n",
    "         epochs=1,\n",
    "         validation_split=0.2)\n",
    "\n",
    "#model.predict(x_test, batch_size=32)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.共享网络层"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "share_embedding = layers.Embedding(1000, 64)\n",
    "\n",
    "input1 = keras.Input(shape=(None,), dtype='int32')\n",
    "input2 = keras.Input(shape=(None,), dtype='int32')\n",
    "\n",
    "feat1 = share_embedding(input1)\n",
    "feat2 = share_embedding(input2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.模型复用"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.applications import VGG16\n",
    "vgg16=VGG16()\n",
    "\n",
    "feature_list = [layer.output for layer in vgg16.layers]\n",
    "feat_ext_model = keras.Model(inputs=vgg16.input, outputs=feature_list)\n",
    "\n",
    "img = np.random.random((1, 224, 224, 3).astype('float32'))\n",
    "ext_features = feat_ext_model(img)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 6.自定义网络层"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import tensorflow as tf\n",
    "# import tensorflow.keras as keras\n",
    "class MyDense(layers.Layer):\n",
    "    def __init__(self, units=32):\n",
    "        super(MyDense, self).__init__()\n",
    "        self.units = units\n",
    "    def build(self, input_shape):\n",
    "        self.w = self.add_weight(shape=(input_shape[-1], self.units),\n",
    "                                 initializer='random_normal',\n",
    "                                 trainable=True)\n",
    "        self.b = self.add_weight(shape=(self.units,),\n",
    "                                 initializer='random_normal',\n",
    "                                 trainable=True)\n",
    "    def call(self, inputs):\n",
    "        return tf.matmul(inputs, self.w) + self.b\n",
    "    \n",
    "    def get_config(self):\n",
    "        return {'units': self.units}\n",
    "    \n",
    "inputs = keras.Input((4,))\n",
    "outputs = MyDense(10)(inputs)\n",
    "model = keras.Model(inputs, outputs)\n",
    "config = model.get_config()\n",
    "new_model = keras.Model.from_config(\n",
    "config, custom_objects={'MyDense':MyDense}\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(32, 8, 32)\n",
      "(32, 8, 32)\n",
      "(1, 10, 32)\n"
     ]
    }
   ],
   "source": [
    "# 在自定义网络层调用其他网络层\n",
    "\n",
    "# 超参\n",
    "time_step = 10\n",
    "batch_size = 32\n",
    "hidden_dim = 32\n",
    "inputs_dim = 5\n",
    "\n",
    "# 网络\n",
    "class MyRnn(layers.Layer):\n",
    "    def __init__(self):\n",
    "        super(MyRnn, self).__init__()\n",
    "        self.hidden_dim = hidden_dim\n",
    "        self.projection1 = layers.Dense(units=hidden_dim, activation='relu')\n",
    "        self.projection2 = layers.Dense(units=hidden_dim, activation='relu')\n",
    "        self.classifier = layers.Dense(1, activation='sigmoid')\n",
    "    def call(self, inputs):\n",
    "        outs = []\n",
    "        states = tf.zeros(shape=[inputs.shape[0], self.hidden_dim])\n",
    "        for t in range(inputs.shape[1]):\n",
    "            x = inputs[:,t,:]\n",
    "            h = self.projection1(x)\n",
    "            y = h + self.projection2(states)\n",
    "            states = y\n",
    "            outs.append(y)\n",
    "        # print(outs)\n",
    "        features = tf.stack(outs, axis=1)\n",
    "        print(features.shape)\n",
    "        return self.classifier(features)\n",
    "\n",
    "# 构建网络\n",
    "inputs = keras.Input(batch_shape=(batch_size, time_step, inputs_dim))\n",
    "x = layers.Conv1D(32, 3)(inputs)\n",
    "print(x.shape)\n",
    "outputs = MyRnn()(x)\n",
    "model = keras.Model(inputs, outputs)\n",
    "\n",
    "\n",
    "rnn_model = MyRnn()\n",
    "_ = rnn_model(tf.zeros((1, 10, 5)))\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"small resnet\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "img (InputLayer)             [(None, 32, 32, 3)]       0         \n",
      "_________________________________________________________________\n",
      "conv2d_8 (Conv2D)            (None, 30, 30, 32)        896       \n",
      "_________________________________________________________________\n",
      "conv2d_9 (Conv2D)            (None, 28, 28, 64)        18496     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_4 (MaxPooling2 (None, 9, 9, 64)          0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 5184)              0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 256)               1327360   \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 10)                2570      \n",
      "=================================================================\n",
      "Total params: 1,349,322\n",
      "Trainable params: 1,349,322\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 简单的卷积神经网络\n",
    "\n",
    "from __future__ import absolute_import, division, print_function\n",
    "import tensorflow as tf\n",
    "import tensorflow.keras as keras\n",
    "import tensorflow.keras.layers as layers\n",
    "inputs = keras.Input(shape=(32,32,3), name='img')\n",
    "h1 = layers.Conv2D(32, 3, activation='relu')(inputs)\n",
    "h2 = layers.Conv2D(64, 3, activation='relu')(h1)\n",
    "h3 = layers.MaxPooling2D(3)(h2)\n",
    "h3 = layers.Flatten()(h3)\n",
    "h4 = layers.Dense(256, activation='relu')(h3)\n",
    "h4 = layers.Dropout(0.5)(h4)\n",
    "outputs = layers.Dense(10, activation='softmax')(h4)\n",
    "\n",
    "model = keras.Model(inputs, outputs, name='small resnet')\n",
    "model.summary()\n",
    "keras.utils.plot_model(model, 'small_resnet_model.png', show_shapes=True)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "keras.layers.GlobalMaxPool1D\n",
    "(x_train, y_train), (x_test, y_test) = keras.datasets.cifar10.load_data()\n",
    "x_train = x_train.astype('float32') / 255\n",
    "x_test = y_train.astype('float32') / 255\n",
    "y_train = keras.utils.to_categorical(y_train, 10)\n",
    "y_test = keras.utils.to_categorical(y_test, 10)\n",
    "\n",
    "model.compile(optimizer=keras.optimizers.RMSprop(1e-3),\n",
    "             loss='categorical_crossentropy',\n",
    "             metrics=['acc'])\n",
    "model.fit(x_train, y_train,\n",
    "         batch_size=64,\n",
    "         epochs=1,\n",
    "         validation_split=0.2)\n",
    "\n",
    "#model.predict(x_test, batch_size=32)\n"
   ]
  }
 ],
 "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}