tf2-keras-key-features

tf2-keras-key-features.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "import random\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### TF1 used to...\n",
    "- `kitchen sink`: need to prepare everything before starting session"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```\n",
    "# params setting\n",
    "    learning_rate = 0.001\n",
    "    epochs = 5\n",
    "    batch_size = 64\n",
    "    n_input = 784\n",
    "    n_classes = 10\n",
    "    n_hidden_layer = 256\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```\n",
    "# weight and biases\n",
    "    weights = {\n",
    "        'hidden_layer': tf.Variable(tf.random.normal([n_input, n_hidden_layer])),\n",
    "        'output': tf.Variable(tf.random.normal([n_hidden_layer, n_classes]))}\n",
    "    biases = {\n",
    "        'hidden_layer': tf.Variable(tf.random.normal([n_hidden_layer])),\n",
    "        'output': tf.Variable(tf.random.normal([n_classes]))}\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```\n",
    "# input\n",
    "    tf.compat.v1.disable_eager_execution()\n",
    "    x = tf.compat.v1.placeholder('float32', [None, 784])\n",
    "    y = tf.compat.v1.placeholder('float32', [None, n_classes])\n",
    "\n",
    "# hidden layer with Relu\n",
    "    layer_1 = tf.add(tf.matmul(x, weights['hidden_layer']), biases['hidden_layer'])\n",
    "    layer_1 = tf.compat.v1.nn.relu(layer_1)\n",
    "    logits = tf.add(tf.matmul(layer_1, weights['output']), biases['output'])\n",
    "\n",
    "# loss and optimizer\n",
    "    cost = tf.reduce_mean(tf.compat.v1.nn.softmax_cross_entropy_with_logits(logits=logits, labels=y))\n",
    "    optimizer = tf.compat.v1.train.GradientDescentOptimizer(learning_rate=learning_rate).minimize(cost)\n",
    "\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```\n",
    "# Sesssion\n",
    "    init = tf.compat.v1.global_variables_initializer()\n",
    "\n",
    "    with tf.compat.v1.Session() as sess:\n",
    "        sess.run(init)\n",
    "        # training\n",
    "        for epoch in range(epochs):\n",
    "            total_batch = int(len(train_dataset) / batch_size)\n",
    "            for i in range(total_batch):\n",
    "                batch_x, batch_y = train_dataset.batch(batch_size)\n",
    "                sess.run(optimizer, feed_dict={x: batch_x, y: batch_y})\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### TF2 does this..\n",
    "- `Eager execution`: no need to create graph first, just execute line by line"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "# data load\n",
    "(x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data()\n",
    "\n",
    "x_train = x_train.reshape(60000, 784).astype('float32') / 255.\n",
    "x_test = x_test.reshape(10000, 784).astype('float32') / 255.\n",
    "\n",
    "y_train = y_train.astype('float32')\n",
    "y_test =y_test.astype('float32')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Sequential API example\n",
    "- staking layer in one direction\n",
    "- only can handle models linear topology\n",
    "\n",
    "```\n",
    "(input: 784-dimensional vectors)\n",
    "       ↧\n",
    "[Dense (64 units, relu activation)]\n",
    "       ↧\n",
    "[Dense (64 units, relu activation)]\n",
    "       ↧\n",
    "[Dense (10 units, softmax activation)]\n",
    "       ↧\n",
    "(output: logits of a probability distribution over 10 classes)\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "# build model; network structure\n",
    "model = tf.keras.Sequential()\n",
    "model.add(layers.Dense(64, activation='relu'))  # hidden layer\n",
    "model.add(layers.Dense(64, activation='relu'))  # if you want add more ..\n",
    "model.add(layers.Dense(10, activation='softmax'))  # output layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "# compile\n",
    "model.compile(optimizer=tf.keras.optimizers.RMSprop(),\n",
    "              loss=keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
    "              metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 60000 samples\n",
      "Epoch 1/3\n",
      "60000/60000 [==============================] - 4s 60us/sample - loss: 1.6061 - accuracy: 0.8734\n",
      "Epoch 2/3\n",
      "60000/60000 [==============================] - 4s 70us/sample - loss: 1.5258 - accuracy: 0.9400\n",
      "Epoch 3/3\n",
      "60000/60000 [==============================] - 5s 80us/sample - loss: 1.5122 - accuracy: 0.9522\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x7f3597ea3550>"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# fit model; train\n",
    "model.fit(x_train, y_train, epochs=3, batch_size=64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10000/10000 [==============================] - 0s 46us/sample - loss: 1.5136 - accuracy: 0.9491\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[1.5136361794114113, 0.9491]"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# evaluate\n",
    "model.evaluate(x_test, y_test, batch_size=64)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Functional API\n",
    "- The Functional API is a way to create models that is more flexible than `Sequential`\n",
    "- It can handle models with **non-linear topology**, models with shared layers, and models with multiple inputs or outputs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "# inputs\n",
    "inputs = keras.Input(shape=(784,))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "# layers, model\n",
    "x = layers.Dense(64, activation='relu')(inputs)\n",
    "x = layers.Dense(64, activation='relu')(x)\n",
    "outputs = layers.Dense(10, activation='softmax')(x)\n",
    "\n",
    "model = keras.Model(inputs=inputs, outputs=outputs, name='mnist_model')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"mnist_model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_2 (InputLayer)         [(None, 784)]             0         \n",
      "_________________________________________________________________\n",
      "dense_14 (Dense)             (None, 64)                50240     \n",
      "_________________________________________________________________\n",
      "dense_15 (Dense)             (None, 64)                4160      \n",
      "_________________________________________________________________\n",
      "dense_16 (Dense)             (None, 10)                650       \n",
      "=================================================================\n",
      "Total params: 55,050\n",
      "Trainable params: 55,050\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keras.utils.plot_model(model, 'mnist_model.png', show_shapes=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "# same process before\n",
    "# compile\n",
    "model.compile(optimizer=tf.keras.optimizers.RMSprop(),\n",
    "              loss=keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
    "              metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 60000 samples\n",
      "Epoch 1/3\n",
      "60000/60000 [==============================] - 5s 77us/sample - loss: 1.6573 - accuracy: 0.8183\n",
      "Epoch 2/3\n",
      "60000/60000 [==============================] - 5s 88us/sample - loss: 1.6059 - accuracy: 0.8574\n",
      "Epoch 3/3\n",
      "60000/60000 [==============================] - 4s 69us/sample - loss: 1.5688 - accuracy: 0.8948\n"
     ]
    }
   ],
   "source": [
    "# fit model; train\n",
    "history = model.fit(x_train, y_train, epochs=3, batch_size=64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'loss': [1.657320616743962, 1.6058963883042336, 1.5687547155896822],\n",
       " 'accuracy': [0.8183166666666667, 0.8574166666666667, 0.8948]}"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "history.history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10000/10000 [==============================] - 1s 92us/sample - loss: 1.5142 - accuracy: 0.9516\n"
     ]
    }
   ],
   "source": [
    "test_scores = model.evaluate(x_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Multi-input, multi-output model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "image_input = keras.Input(shape=(32, 32, 3), name='img_input')\n",
    "timeseries_input = keras.Input(shape=(None, 10), name='ts_input')\n",
    "\n",
    "x1 = layers.Conv2D(3, 3)(image_input)\n",
    "x1 = layers.GlobalMaxPooling2D()(x1)\n",
    "\n",
    "x2 = layers.Conv1D(3, 3)(timeseries_input)\n",
    "x2 = layers.GlobalMaxPooling1D()(x2)\n",
    "\n",
    "x = layers.concatenate([x1, x2])\n",
    "\n",
    "score_output = layers.Dense(1, name='score_output')(x)\n",
    "class_output = layers.Dense(5, name='class_output')(x)\n",
    "\n",
    "# two different inputs & outputs\n",
    "model = keras.Model(inputs=[image_input, timeseries_input],\n",
    "                    outputs=[score_output, class_output])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keras.utils.plot_model(model, 'multi_input_and_output_model.png', show_shapes=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "# two different losses & metrics for different inputs & outputs\n",
    "model.compile(\n",
    "    optimizer=keras.optimizers.RMSprop(1e-3),\n",
    "    loss=[keras.losses.MeanSquaredError(),\n",
    "          keras.losses.CategoricalCrossentropy(from_logits=True)],\n",
    "    metrics=[[keras.metrics.MeanAbsolutePercentageError(),\n",
    "              keras.metrics.MeanAbsoluteError()],\n",
    "             [keras.metrics.CategoricalAccuracy()]])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Autoencoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "scrolled": false
   },
   "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_1 (Conv2D)            (None, 26, 26, 16)        160       \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 24, 24, 32)        4640      \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 8, 8, 32)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 6, 6, 32)          9248      \n",
      "_________________________________________________________________\n",
      "conv2d_4 (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_1 (Conv2D)            (None, 26, 26, 16)        160       \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 24, 24, 32)        4640      \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 8, 8, 32)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 6, 6, 32)          9248      \n",
      "_________________________________________________________________\n",
      "conv2d_4 (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": [
    "encoder_input = keras.Input(shape=(28, 28, 1), name='img')\n",
    "x = layers.Conv2D(16, 3, activation='relu')(encoder_input)\n",
    "x = layers.Conv2D(32, 3, activation='relu')(x)\n",
    "x = layers.MaxPooling2D(3)(x)\n",
    "x = layers.Conv2D(32, 3, activation='relu')(x)\n",
    "x = layers.Conv2D(16, 3, activation='relu')(x)\n",
    "encoder_output = layers.GlobalMaxPooling2D()(x)\n",
    "\n",
    "encoder = keras.Model(encoder_input, encoder_output, name='encoder')\n",
    "encoder.summary()\n",
    "\n",
    "x = layers.Reshape((4, 4, 1))(encoder_output)\n",
    "x = layers.Conv2DTranspose(16, 3, activation='relu')(x)\n",
    "x = layers.Conv2DTranspose(32, 3, activation='relu')(x)\n",
    "x = layers.UpSampling2D(3)(x)\n",
    "x = layers.Conv2DTranspose(16, 3, activation='relu')(x)\n",
    "decoder_output = layers.Conv2DTranspose(1, 3, activation='relu')(x)\n",
    "\n",
    "autoencoder = keras.Model(encoder_input, decoder_output, name='autoencoder')\n",
    "autoencoder.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keras.utils.plot_model(autoencoder, 'autoencoder.png', show_shapes=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![](autoencoder_structure.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Extending the API by writing custom layers - Subclassing\n",
    "tf.keras has a wide range of built-in layers. Here are a few examples:\n",
    "- Convolutional layers: `Conv1D`, `Conv2D`, `Conv3D`, `Conv2DTranpose`, etc.\n",
    "- Pooling layers: `MaxPooling1D`, `MaxPooling2D`, `MaxPooling3D`, `AveragePooling1D`, etc.\n",
    "- RNN layers: `GRU`, `LSTM`, `ConvLSTM2D`, etc.\n",
    "- `BatchNormalization`, `Dropout`, `Embedding`, etc.\n",
    "\n",
    "If you don't find what you need, it's easy to extend the API by creating your own layers.\n",
    "\n",
    "All layers subclass the `Layer` class and implement:\n",
    "- A `call` method, that specifies the computation done by the layer.\n",
    "- A `build` method, that creates the weights of the layer (note that this is just a style convention; you could create weights in `__init__` as well).\n",
    "\n",
    "Here's a simple implementation of a `Dense` layer:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "class CustomDense(layers.Layer):\n",
    "\n",
    "  def __init__(self, units=32):\n",
    "    super(CustomDense, self).__init__()\n",
    "    self.units = units\n",
    "\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",
    "\n",
    "  def call(self, inputs):\n",
    "    return tf.matmul(inputs, self.w) + self.b\n",
    "\n",
    "inputs = keras.Input((4,))\n",
    "outputs = CustomDense(10)(inputs)\n",
    "\n",
    "model = keras.Model(inputs, outputs)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### And there're many things...."
   ]
  }
 ],
 "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}