{
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "b518b04cbfe0"
},
"source": [
"##### Copyright 2020 The TensorFlow Authors."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"cellView": "form",
"execution": {
"iopub.execute_input": "2022-12-14T21:33:01.723714Z",
"iopub.status.busy": "2022-12-14T21:33:01.723181Z",
"iopub.status.idle": "2022-12-14T21:33:01.726810Z",
"shell.execute_reply": "2022-12-14T21:33:01.726278Z"
},
"id": "906e07f6e562"
},
"outputs": [],
"source": [
"#@title Licensed under the Apache License, Version 2.0 (the \"License\");\n",
"# you may not use this file except in compliance with the License.\n",
"# You may obtain a copy of the License at\n",
"#\n",
"# https://www.apache.org/licenses/LICENSE-2.0\n",
"#\n",
"# Unless required by applicable law or agreed to in writing, software\n",
"# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
"# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
"# See the License for the specific language governing permissions and\n",
"# limitations under the License."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "e2d97c7e31aa"
},
"source": [
"# 通过子类化创建新的层和模型"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "4e352274064f"
},
"source": [
""
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "8d4ac441b1fc"
},
"source": [
"## 设置"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:01.730044Z",
"iopub.status.busy": "2022-12-14T21:33:01.729610Z",
"iopub.status.idle": "2022-12-14T21:33:03.631073Z",
"shell.execute_reply": "2022-12-14T21:33:03.630382Z"
},
"id": "4e7dce39dd1d"
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"2022-12-14 21:33:02.655873: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory\n",
"2022-12-14 21:33:02.655988: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory\n",
"2022-12-14 21:33:02.655999: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.\n"
]
}
],
"source": [
"import tensorflow as tf\n",
"from tensorflow import keras"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "7b363673d96c"
},
"source": [
"## `Layer` 类：状态（权重）和部分计算的组合\n",
"\n",
"Keras 的一个中心抽象是 `Layer` 类。层封装了状态（层的“权重”）和从输入到输出的转换（“调用”，即层的前向传递）。\n",
"\n",
"下面是一个密集连接的层。它具有一个状态：变量 `w` 和 `b`。"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:03.635235Z",
"iopub.status.busy": "2022-12-14T21:33:03.634606Z",
"iopub.status.idle": "2022-12-14T21:33:03.639685Z",
"shell.execute_reply": "2022-12-14T21:33:03.639031Z"
},
"id": "59b8317dbd3c"
},
"outputs": [],
"source": [
"class Linear(keras.layers.Layer):\n",
" def __init__(self, units=32, input_dim=32):\n",
" super(Linear, self).__init__()\n",
" w_init = tf.random_normal_initializer()\n",
" self.w = tf.Variable(\n",
" initial_value=w_init(shape=(input_dim, units), dtype=\"float32\"),\n",
" trainable=True,\n",
" )\n",
" b_init = tf.zeros_initializer()\n",
" self.b = tf.Variable(\n",
" initial_value=b_init(shape=(units,), dtype=\"float32\"), trainable=True\n",
" )\n",
"\n",
" def call(self, inputs):\n",
" return tf.matmul(inputs, self.w) + self.b\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "dac8fb03a642"
},
"source": [
"您可以在某些张量输入上通过调用来使用层，这一点很像 Python 函数。"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:03.642744Z",
"iopub.status.busy": "2022-12-14T21:33:03.642291Z",
"iopub.status.idle": "2022-12-14T21:33:07.355937Z",
"shell.execute_reply": "2022-12-14T21:33:07.355209Z"
},
"id": "cdcd15d5e68a"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"tf.Tensor(\n",
"[[-0.00204108 -0.04129961 -0.01273937 -0.1288575 ]\n",
" [-0.00204108 -0.04129961 -0.01273937 -0.1288575 ]], shape=(2, 4), dtype=float32)\n"
]
}
],
"source": [
"x = tf.ones((2, 2))\n",
"linear_layer = Linear(4, 2)\n",
"y = linear_layer(x)\n",
"print(y)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "382960020a56"
},
"source": [
"请注意，权重 `w` 和 `b` 在被设置为层特性后会由层自动跟踪："
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.359521Z",
"iopub.status.busy": "2022-12-14T21:33:07.358990Z",
"iopub.status.idle": "2022-12-14T21:33:07.362526Z",
"shell.execute_reply": "2022-12-14T21:33:07.361935Z"
},
"id": "d3d875af9465"
},
"outputs": [],
"source": [
"assert linear_layer.weights == [linear_layer.w, linear_layer.b]"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "ec9d72aa7538"
},
"source": [
"请注意，您还可以使用一种更加快捷的方式为层添加权重：`add_weight()` 方法："
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.365588Z",
"iopub.status.busy": "2022-12-14T21:33:07.365147Z",
"iopub.status.idle": "2022-12-14T21:33:07.380503Z",
"shell.execute_reply": "2022-12-14T21:33:07.379875Z"
},
"id": "168548eba841"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"tf.Tensor(\n",
"[[-0.11710902 0.10924062 0.02172409 0.02696215]\n",
" [-0.11710902 0.10924062 0.02172409 0.02696215]], shape=(2, 4), dtype=float32)\n"
]
}
],
"source": [
"class Linear(keras.layers.Layer):\n",
" def __init__(self, units=32, input_dim=32):\n",
" super(Linear, self).__init__()\n",
" self.w = self.add_weight(\n",
" shape=(input_dim, units), initializer=\"random_normal\", trainable=True\n",
" )\n",
" self.b = self.add_weight(shape=(units,), initializer=\"zeros\", trainable=True)\n",
"\n",
" def call(self, inputs):\n",
" return tf.matmul(inputs, self.w) + self.b\n",
"\n",
"\n",
"x = tf.ones((2, 2))\n",
"linear_layer = Linear(4, 2)\n",
"y = linear_layer(x)\n",
"print(y)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "070ea9b4db6c"
},
"source": [
"## 层可以具有不可训练权重\n",
"\n",
"除了可训练权重外，您还可以向层添加不可训练权重。训练层时，不必在反向传播期间考虑此类权重。\n",
"\n",
"以下是添加和使用不可训练权重的方式："
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.383536Z",
"iopub.status.busy": "2022-12-14T21:33:07.383179Z",
"iopub.status.idle": "2022-12-14T21:33:07.396901Z",
"shell.execute_reply": "2022-12-14T21:33:07.396312Z"
},
"id": "7c4cb404145f"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2. 2.]\n",
"[4. 4.]\n"
]
}
],
"source": [
"class ComputeSum(keras.layers.Layer):\n",
" def __init__(self, input_dim):\n",
" super(ComputeSum, self).__init__()\n",
" self.total = tf.Variable(initial_value=tf.zeros((input_dim,)), trainable=False)\n",
"\n",
" def call(self, inputs):\n",
" self.total.assign_add(tf.reduce_sum(inputs, axis=0))\n",
" return self.total\n",
"\n",
"\n",
"x = tf.ones((2, 2))\n",
"my_sum = ComputeSum(2)\n",
"y = my_sum(x)\n",
"print(y.numpy())\n",
"y = my_sum(x)\n",
"print(y.numpy())"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "40f5b74d3d87"
},
"source": [
"它是 `layer.weights` 的一部分，但被归类为不可训练权重："
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.399994Z",
"iopub.status.busy": "2022-12-14T21:33:07.399392Z",
"iopub.status.idle": "2022-12-14T21:33:07.403097Z",
"shell.execute_reply": "2022-12-14T21:33:07.402551Z"
},
"id": "3d4db4ef4fa4"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"weights: 1\n",
"non-trainable weights: 1\n",
"trainable_weights: []\n"
]
}
],
"source": [
"print(\"weights:\", len(my_sum.weights))\n",
"print(\"non-trainable weights:\", len(my_sum.non_trainable_weights))\n",
"\n",
"# It's not included in the trainable weights:\n",
"print(\"trainable_weights:\", my_sum.trainable_weights)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "fe6942aff7c6"
},
"source": [
"## 最佳做法：将权重创建推迟到得知输入的形状之后\n",
"\n",
"上面的 `Linear` 层接受了一个 `input_dim` 参数，用于计算 `__init__()` 中权重 `w` 和 `b` 的形状："
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.405963Z",
"iopub.status.busy": "2022-12-14T21:33:07.405504Z",
"iopub.status.idle": "2022-12-14T21:33:07.409760Z",
"shell.execute_reply": "2022-12-14T21:33:07.409187Z"
},
"id": "275b68d5ea9f"
},
"outputs": [],
"source": [
"class Linear(keras.layers.Layer):\n",
" def __init__(self, units=32, input_dim=32):\n",
" super(Linear, self).__init__()\n",
" self.w = self.add_weight(\n",
" shape=(input_dim, units), initializer=\"random_normal\", trainable=True\n",
" )\n",
" self.b = self.add_weight(shape=(units,), initializer=\"zeros\", trainable=True)\n",
"\n",
" def call(self, inputs):\n",
" return tf.matmul(inputs, self.w) + self.b\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "5ebcacebb348"
},
"source": [
"在许多情况下，您可能事先不知道输入的大小，并希望在得知该值时（对层进行实例化后的某个时间）再延迟创建权重。\n",
"\n",
"在 Keras API 中，我们建议您在层的 `build(self, inputs_shape)` 方法中创建层权重。如下所示："
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.412817Z",
"iopub.status.busy": "2022-12-14T21:33:07.412341Z",
"iopub.status.idle": "2022-12-14T21:33:07.416716Z",
"shell.execute_reply": "2022-12-14T21:33:07.416165Z"
},
"id": "118c899f427e"
},
"outputs": [],
"source": [
"class Linear(keras.layers.Layer):\n",
" def __init__(self, units=32):\n",
" super(Linear, self).__init__()\n",
" self.units = units\n",
"\n",
" def build(self, input_shape):\n",
" self.w = self.add_weight(\n",
" shape=(input_shape[-1], self.units),\n",
" initializer=\"random_normal\",\n",
" trainable=True,\n",
" )\n",
" self.b = self.add_weight(\n",
" shape=(self.units,), initializer=\"random_normal\", trainable=True\n",
" )\n",
"\n",
" def call(self, inputs):\n",
" return tf.matmul(inputs, self.w) + self.b\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "78061e0583c6"
},
"source": [
"层的 `__call__()` 方法将在首次调用时自动运行构建。现在，您有了一个延迟并因此更易使用的层："
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.419676Z",
"iopub.status.busy": "2022-12-14T21:33:07.419117Z",
"iopub.status.idle": "2022-12-14T21:33:07.432160Z",
"shell.execute_reply": "2022-12-14T21:33:07.431592Z"
},
"id": "0697afb97bc1"
},
"outputs": [],
"source": [
"# At instantiation, we don't know on what inputs this is going to get called\n",
"linear_layer = Linear(32)\n",
"\n",
"# The layer's weights are created dynamically the first time the layer is called\n",
"y = linear_layer(x)\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "51b81f42b466"
},
"source": [
"如上所示单独实现 `build()` 很好地将只创建一次权重与在每次调用时使用权重分开。但是，对于一些高级自定义层，将状态创建和计算分开可能变得不切实际。层实现器可以将权重创建推迟到第一个 `__call__()`，但需要注意，后面的调用会使用相同的权重。此外，由于 `__call__()` 很可能是第一次在 `tf.function` 中执行，在 `__call__()` 中发生的任何变量创建都应当封装在 `tf.init_scope` 中。"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "0b7a45f57610"
},
"source": [
"## 层可递归组合\n",
"\n",
"如果将层实例分配为另一个层的特性，则外部层将开始跟踪内部层创建的权重。\n",
"\n",
"我们建议在 `__init__()` 方法中创建此类子层，并将其留给第一个 `__call__()` 以触发构建它们的权重。"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.435306Z",
"iopub.status.busy": "2022-12-14T21:33:07.434837Z",
"iopub.status.idle": "2022-12-14T21:33:07.465534Z",
"shell.execute_reply": "2022-12-14T21:33:07.464945Z"
},
"id": "1aaaf82ab8ce"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"weights: 6\n",
"trainable weights: 6\n"
]
}
],
"source": [
"class MLPBlock(keras.layers.Layer):\n",
" def __init__(self):\n",
" super(MLPBlock, self).__init__()\n",
" self.linear_1 = Linear(32)\n",
" self.linear_2 = Linear(32)\n",
" self.linear_3 = Linear(1)\n",
"\n",
" def call(self, inputs):\n",
" x = self.linear_1(inputs)\n",
" x = tf.nn.relu(x)\n",
" x = self.linear_2(x)\n",
" x = tf.nn.relu(x)\n",
" return self.linear_3(x)\n",
"\n",
"\n",
"mlp = MLPBlock()\n",
"y = mlp(tf.ones(shape=(3, 64))) # The first call to the `mlp` will create the weights\n",
"print(\"weights:\", len(mlp.weights))\n",
"print(\"trainable weights:\", len(mlp.trainable_weights))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "2bf11b296bd2"
},
"source": [
"## `add_loss()` 方法\n",
"\n",
"在编写层的 `call()` 方法时，您可以在编写训练循环时创建想要稍后使用的损失张量。这可以通过调用 `self.add_loss(value)` 来实现："
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.468538Z",
"iopub.status.busy": "2022-12-14T21:33:07.468054Z",
"iopub.status.idle": "2022-12-14T21:33:07.472005Z",
"shell.execute_reply": "2022-12-14T21:33:07.471390Z"
},
"id": "ba2782dc0879"
},
"outputs": [],
"source": [
"# A layer that creates an activity regularization loss\n",
"class ActivityRegularizationLayer(keras.layers.Layer):\n",
" def __init__(self, rate=1e-2):\n",
" super(ActivityRegularizationLayer, self).__init__()\n",
" self.rate = rate\n",
"\n",
" def call(self, inputs):\n",
" self.add_loss(self.rate * tf.reduce_sum(inputs))\n",
" return inputs\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "a883b230a9e9"
},
"source": [
"这些损失（包括由任何内部层创建的损失）可通过 `layer.losses` 取回。此属性会在每个 `__call__()` 开始时重置到顶层，因此 `layer.losses` 始终包含在上一次前向传递过程中创建的损失值。"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.474841Z",
"iopub.status.busy": "2022-12-14T21:33:07.474376Z",
"iopub.status.idle": "2022-12-14T21:33:07.483815Z",
"shell.execute_reply": "2022-12-14T21:33:07.483186Z"
},
"id": "b56d223a30cd"
},
"outputs": [],
"source": [
"class OuterLayer(keras.layers.Layer):\n",
" def __init__(self):\n",
" super(OuterLayer, self).__init__()\n",
" self.activity_reg = ActivityRegularizationLayer(1e-2)\n",
"\n",
" def call(self, inputs):\n",
" return self.activity_reg(inputs)\n",
"\n",
"\n",
"layer = OuterLayer()\n",
"assert len(layer.losses) == 0 # No losses yet since the layer has never been called\n",
"\n",
"_ = layer(tf.zeros(1, 1))\n",
"assert len(layer.losses) == 1 # We created one loss value\n",
"\n",
"# `layer.losses` gets reset at the start of each __call__\n",
"_ = layer(tf.zeros(1, 1))\n",
"assert len(layer.losses) == 1 # This is the loss created during the call above"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "0809dec680ff"
},
"source": [
"此外，`loss` 属性还包含为任何内部层的权重创建的正则化损失："
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.486973Z",
"iopub.status.busy": "2022-12-14T21:33:07.486377Z",
"iopub.status.idle": "2022-12-14T21:33:07.505805Z",
"shell.execute_reply": "2022-12-14T21:33:07.505220Z"
},
"id": "41016153e983"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[]\n"
]
}
],
"source": [
"class OuterLayerWithKernelRegularizer(keras.layers.Layer):\n",
" def __init__(self):\n",
" super(OuterLayerWithKernelRegularizer, self).__init__()\n",
" self.dense = keras.layers.Dense(\n",
" 32, kernel_regularizer=tf.keras.regularizers.l2(1e-3)\n",
" )\n",
"\n",
" def call(self, inputs):\n",
" return self.dense(inputs)\n",
"\n",
"\n",
"layer = OuterLayerWithKernelRegularizer()\n",
"_ = layer(tf.zeros((1, 1)))\n",
"\n",
"# This is `1e-3 * sum(layer.dense.kernel ** 2)`,\n",
"# created by the `kernel_regularizer` above.\n",
"print(layer.losses)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "589465e06e4f"
},
"source": [
"在编写训练循环时应考虑这些损失，如下所示：\n",
"\n",
"```python\n",
"# Instantiate an optimizer.\n",
"optimizer = tf.keras.optimizers.SGD(learning_rate=1e-3)\n",
"loss_fn = keras.losses.SparseCategoricalCrossentropy(from_logits=True)\n",
"\n",
"# Iterate over the batches of a dataset.\n",
"for x_batch_train, y_batch_train in train_dataset:\n",
" with tf.GradientTape() as tape:\n",
" logits = layer(x_batch_train) # Logits for this minibatch\n",
" # Loss value for this minibatch\n",
" loss_value = loss_fn(y_batch_train, logits)\n",
" # Add extra losses created during this forward pass:\n",
" loss_value += sum(model.losses)\n",
"\n",
" grads = tape.gradient(loss_value, model.trainable_weights)\n",
" optimizer.apply_gradients(zip(grads, model.trainable_weights))\n",
"```"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "7fb41ca8c3b0"
},
"source": [
"有关编写训练循环的详细指南，请参阅[从头开始编写训练循环](https://tensorflow.google.cn/guide/keras/writing_a_training_loop_from_scratch/)指南。\n",
"\n",
"这些损失还可以无缝使用 `fit()`（它们会自动求和并添加到主损失中，如果有）："
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.508924Z",
"iopub.status.busy": "2022-12-14T21:33:07.508461Z",
"iopub.status.idle": "2022-12-14T21:33:07.903278Z",
"shell.execute_reply": "2022-12-14T21:33:07.902618Z"
},
"id": "769bc6612ebf"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\r",
"1/1 [==============================] - ETA: 0s - loss: 0.2993"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1/1 [==============================] - 0s 178ms/step - loss: 0.2993\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\r",
"1/1 [==============================] - ETA: 0s - loss: 0.0257"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1/1 [==============================] - 0s 33ms/step - loss: 0.0257\n"
]
},
{
"data": {
"text/plain": [
""
]
},
"execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import numpy as np\n",
"\n",
"inputs = keras.Input(shape=(3,))\n",
"outputs = ActivityRegularizationLayer()(inputs)\n",
"model = keras.Model(inputs, outputs)\n",
"\n",
"# If there is a loss passed in `compile`, the regularization\n",
"# losses get added to it\n",
"model.compile(optimizer=\"adam\", loss=\"mse\")\n",
"model.fit(np.random.random((2, 3)), np.random.random((2, 3)))\n",
"\n",
"# It's also possible not to pass any loss in `compile`,\n",
"# since the model already has a loss to minimize, via the `add_loss`\n",
"# call during the forward pass!\n",
"model.compile(optimizer=\"adam\")\n",
"model.fit(np.random.random((2, 3)), np.random.random((2, 3)))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "149c71e442bb"
},
"source": [
"## `add_metric()` 方法\n",
"\n",
"与 `add_loss()` 类似，层还具有 `add_metric()` 方法，用于在训练过程中跟踪数量的移动平均值。\n",
"\n",
"请思考下面的 \"logistic endpoint\" 层。它将预测和目标作为输入，计算通过 `add_loss()` 跟踪的损失，并计算通过 `add_metric()` 跟踪的准确率标量。"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.906964Z",
"iopub.status.busy": "2022-12-14T21:33:07.906335Z",
"iopub.status.idle": "2022-12-14T21:33:07.911715Z",
"shell.execute_reply": "2022-12-14T21:33:07.911000Z"
},
"id": "bfb2df515096"
},
"outputs": [],
"source": [
"class LogisticEndpoint(keras.layers.Layer):\n",
" def __init__(self, name=None):\n",
" super(LogisticEndpoint, self).__init__(name=name)\n",
" self.loss_fn = keras.losses.BinaryCrossentropy(from_logits=True)\n",
" self.accuracy_fn = keras.metrics.BinaryAccuracy()\n",
"\n",
" def call(self, targets, logits, sample_weights=None):\n",
" # Compute the training-time loss value and add it\n",
" # to the layer using `self.add_loss()`.\n",
" loss = self.loss_fn(targets, logits, sample_weights)\n",
" self.add_loss(loss)\n",
"\n",
" # Log accuracy as a metric and add it\n",
" # to the layer using `self.add_metric()`.\n",
" acc = self.accuracy_fn(targets, logits, sample_weights)\n",
" self.add_metric(acc, name=\"accuracy\")\n",
"\n",
" # Return the inference-time prediction tensor (for `.predict()`).\n",
" return tf.nn.softmax(logits)\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "e68f88373800"
},
"source": [
"可通过 `layer.metrics` 访问以这种方式跟踪的指标："
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.915088Z",
"iopub.status.busy": "2022-12-14T21:33:07.914536Z",
"iopub.status.idle": "2022-12-14T21:33:07.944032Z",
"shell.execute_reply": "2022-12-14T21:33:07.943401Z"
},
"id": "1834d74450b6"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"layer.metrics: []\n",
"current accuracy value: 1.0\n"
]
}
],
"source": [
"layer = LogisticEndpoint()\n",
"\n",
"targets = tf.ones((2, 2))\n",
"logits = tf.ones((2, 2))\n",
"y = layer(targets, logits)\n",
"\n",
"print(\"layer.metrics:\", layer.metrics)\n",
"print(\"current accuracy value:\", float(layer.metrics[0].result()))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "546cfbd4ea05"
},
"source": [
"和 `add_loss()` 一样，这些指标也是通过 `fit()` 跟踪的："
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:07.947477Z",
"iopub.status.busy": "2022-12-14T21:33:07.946882Z",
"iopub.status.idle": "2022-12-14T21:33:08.692704Z",
"shell.execute_reply": "2022-12-14T21:33:08.691946Z"
},
"id": "f5e74cb4da34"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\r",
"1/1 [==============================] - ETA: 0s - loss: 0.9536 - binary_accuracy: 0.0000e+00"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1/1 [==============================] - 1s 594ms/step - loss: 0.9536 - binary_accuracy: 0.0000e+00\n"
]
},
{
"data": {
"text/plain": [
""
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"inputs = keras.Input(shape=(3,), name=\"inputs\")\n",
"targets = keras.Input(shape=(10,), name=\"targets\")\n",
"logits = keras.layers.Dense(10)(inputs)\n",
"predictions = LogisticEndpoint(name=\"predictions\")(logits, targets)\n",
"\n",
"model = keras.Model(inputs=[inputs, targets], outputs=predictions)\n",
"model.compile(optimizer=\"adam\")\n",
"\n",
"data = {\n",
" \"inputs\": np.random.random((3, 3)),\n",
" \"targets\": np.random.random((3, 10)),\n",
"}\n",
"model.fit(data)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "4012fa8683e5"
},
"source": [
"## 可选择在层上启用序列化\n",
"\n",
"如果需要将自定义层作为[函数式模型](https://tensorflow.google.cn/guide/keras/functional/)的一部分进行序列化，您可以选择实现 `get_config()` 方法："
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:08.696665Z",
"iopub.status.busy": "2022-12-14T21:33:08.696007Z",
"iopub.status.idle": "2022-12-14T21:33:08.703046Z",
"shell.execute_reply": "2022-12-14T21:33:08.702454Z"
},
"id": "0a720cbd5f54"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"{'units': 64}\n"
]
}
],
"source": [
"class Linear(keras.layers.Layer):\n",
" def __init__(self, units=32):\n",
" super(Linear, self).__init__()\n",
" self.units = units\n",
"\n",
" def build(self, input_shape):\n",
" self.w = self.add_weight(\n",
" shape=(input_shape[-1], self.units),\n",
" initializer=\"random_normal\",\n",
" trainable=True,\n",
" )\n",
" self.b = self.add_weight(\n",
" shape=(self.units,), initializer=\"random_normal\", trainable=True\n",
" )\n",
"\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",
"\n",
"# Now you can recreate the layer from its config:\n",
"layer = Linear(64)\n",
"config = layer.get_config()\n",
"print(config)\n",
"new_layer = Linear.from_config(config)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "1b43aad6c145"
},
"source": [
"请注意，基础 `Layer` 类的 `__init__()` 方法会接受一些关键字参数，尤其是 `name` 和 `dtype`。最好将这些参数传递给 `__init__()` 中的父类，并将其包含在层配置中："
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:08.706397Z",
"iopub.status.busy": "2022-12-14T21:33:08.705869Z",
"iopub.status.idle": "2022-12-14T21:33:08.712995Z",
"shell.execute_reply": "2022-12-14T21:33:08.712328Z"
},
"id": "0cbad8a6e6cd"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"{'name': 'linear_8', 'trainable': True, 'dtype': 'float32', 'units': 64}\n"
]
}
],
"source": [
"class Linear(keras.layers.Layer):\n",
" def __init__(self, units=32, **kwargs):\n",
" super(Linear, self).__init__(**kwargs)\n",
" self.units = units\n",
"\n",
" def build(self, input_shape):\n",
" self.w = self.add_weight(\n",
" shape=(input_shape[-1], self.units),\n",
" initializer=\"random_normal\",\n",
" trainable=True,\n",
" )\n",
" self.b = self.add_weight(\n",
" shape=(self.units,), initializer=\"random_normal\", trainable=True\n",
" )\n",
"\n",
" def call(self, inputs):\n",
" return tf.matmul(inputs, self.w) + self.b\n",
"\n",
" def get_config(self):\n",
" config = super(Linear, self).get_config()\n",
" config.update({\"units\": self.units})\n",
" return config\n",
"\n",
"\n",
"layer = Linear(64)\n",
"config = layer.get_config()\n",
"print(config)\n",
"new_layer = Linear.from_config(config)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "2421f80b5b86"
},
"source": [
"如果根据层的配置对层进行反序列化时需要更大的灵活性，还可以重写 `from_config()` 类方法。下面是 `from_config()` 的基础实现：\n",
"\n",
"```python\n",
"def from_config(cls, config):\n",
" return cls(**config)\n",
"```\n",
"\n",
"要详细了解序列化和保存，请参阅完整的[保存和序列化模型](https://tensorflow.google.cn/guide/keras/save_and_serialize/)指南。"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "3d7e2304a047"
},
"source": [
"## `call()` 方法中的特权 `training` 参数\n",
"\n",
"某些层，尤其是 `BatchNormalization` 层和 `Dropout` 层，在训练和推断期间具有不同的行为。对于此类层，标准做法是在 `call()` 方法中公开 `training`（布尔）参数。\n",
"\n",
"通过在 `call()` 中公开此参数，可以启用内置的训练和评估循环（例如 `fit()`）以在训练和推断中正确使用层。"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:08.716180Z",
"iopub.status.busy": "2022-12-14T21:33:08.715648Z",
"iopub.status.idle": "2022-12-14T21:33:08.719652Z",
"shell.execute_reply": "2022-12-14T21:33:08.718987Z"
},
"id": "a169812c2c00"
},
"outputs": [],
"source": [
"class CustomDropout(keras.layers.Layer):\n",
" def __init__(self, rate, **kwargs):\n",
" super(CustomDropout, self).__init__(**kwargs)\n",
" self.rate = rate\n",
"\n",
" def call(self, inputs, training=None):\n",
" if training:\n",
" return tf.nn.dropout(inputs, rate=self.rate)\n",
" return inputs\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "9e1482c9f010"
},
"source": [
"## `call()` 方法中的特权 `mask` 参数\n",
"\n",
"`call()` 支持的另一个特权参数是 `mask` 参数。\n",
"\n",
"它会出现在所有 Keras RNN 层中。掩码是布尔张量（在输入中每个时间步骤对应一个布尔值），用于在处理时间序列数据时跳过某些输入时间步骤。\n",
"\n",
"当先前的层生成掩码时，Keras 会自动将正确的 `mask` 参数传递给 `__call__()`（针对支持它的层）。掩码生成层是配置了 `mask_zero=True` 的 `Embedding` 层和 `Masking` 层。\n",
"\n",
"要详细了解遮盖以及如何编写启用遮盖的层，请查看[了解填充和遮盖](https://tensorflow.google.cn/guide/keras/masking_and_padding/)指南。"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "344110f9e134"
},
"source": [
"## `Model` 类\n",
"\n",
"通常，您会使用 `Layer` 类来定义内部计算块，并使用 `Model` 类来定义外部模型，即您将训练的对象。\n",
"\n",
"例如，在 ResNet50 模型中，您会有几个子类化 `Layer` 的 ResNet 块，以及一个包含整个 ResNet50 网络的 `Model`。\n",
"\n",
"`Model` 类具有与 `Layer` 相同的 API，但有如下区别：\n",
"\n",
"- 它会公开内置训练、评估和预测循环（`model.fit()`、`model.evaluate()`、`model.predict()`）。\n",
"- 它会通过 `model.layers` 属性公开其内部层的列表。\n",
"- 它会公开保存和序列化 API（`save()`、`save_weights()`…）\n",
"\n",
"实际上，`Layer` 类对应于我们在文献中所称的“层”（如“卷积层”或“循环层”）或“块”（如“ResNet 块”或“Inception 块”）。\n",
"\n",
"同时，`Model` 类对应于文献中所称的“模型”（如“深度学习模型”）或“网络”（如“深度神经网络”）。\n",
"\n",
"因此，如果您想知道“我应该用 `Layer` 类还是 `Model` 类？”，请问自己：我是否需要在它上面调用 `fit()`？我是否需要在它上面调用 `save()`？如果是，则使用 `Model`。如果不是（要么因为您的类只是更大系统中的一个块，要么因为您正在自己编写训练和保存代码），则使用 `Layer`。\n",
"\n",
"例如，我们可以使用上面的 mini-resnet 示例，用它来构建一个 `Model`，该模型可以通过 `fit()` 进行训练，并通过 `save_weights()` 进行保存："
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "09caa642b72e"
},
"source": [
"```python\n",
"class ResNet(tf.keras.Model):\n",
"\n",
" def __init__(self, num_classes=1000):\n",
" super(ResNet, self).__init__()\n",
" self.block_1 = ResNetBlock()\n",
" self.block_2 = ResNetBlock()\n",
" self.global_pool = layers.GlobalAveragePooling2D()\n",
" self.classifier = Dense(num_classes)\n",
"\n",
" def call(self, inputs):\n",
" x = self.block_1(inputs)\n",
" x = self.block_2(x)\n",
" x = self.global_pool(x)\n",
" return self.classifier(x)\n",
"\n",
"\n",
"resnet = ResNet()\n",
"dataset = ...\n",
"resnet.fit(dataset, epochs=10)\n",
"resnet.save(filepath)\n",
"```"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "a2e32d225a1b"
},
"source": [
"## 汇总：端到端示例\n",
"\n",
"到目前为止，您已学习以下内容：\n",
"\n",
"- `Layer` 封装了状态（在 `__init__()` 或 `build()` 中创建）和一些计算（在 `call()` 中定义）。\n",
"- 层可以递归嵌套以创建新的更大的计算块。\n",
"- 层可以通过 `add_loss()` 和 `add_metric()` 创建并跟踪损失（通常是正则化损失）以及指标。\n",
"- 您要训练的外部容器是 `Model`。`Model` 就像 `Layer`，但是添加了训练和序列化实用工具。\n",
"\n",
"让我们将这些内容全部汇总到一个端到端示例：我们将实现一个变分自动编码器 (VAE)，并用 MNIST 数字对其进行训练。\n",
"\n",
"我们的 VAE 将是 `Model` 的一个子类，它是作为子类化 `Layer` 的嵌套组合层进行构建的。它将具有正则化损失（KL 散度）。"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:08.723152Z",
"iopub.status.busy": "2022-12-14T21:33:08.722610Z",
"iopub.status.idle": "2022-12-14T21:33:08.733053Z",
"shell.execute_reply": "2022-12-14T21:33:08.732469Z"
},
"id": "56aaae7af872"
},
"outputs": [],
"source": [
"from tensorflow.keras import layers\n",
"\n",
"\n",
"class Sampling(layers.Layer):\n",
" \"\"\"Uses (z_mean, z_log_var) to sample z, the vector encoding a digit.\"\"\"\n",
"\n",
" def call(self, inputs):\n",
" z_mean, z_log_var = inputs\n",
" batch = tf.shape(z_mean)[0]\n",
" dim = tf.shape(z_mean)[1]\n",
" epsilon = tf.keras.backend.random_normal(shape=(batch, dim))\n",
" return z_mean + tf.exp(0.5 * z_log_var) * epsilon\n",
"\n",
"\n",
"class Encoder(layers.Layer):\n",
" \"\"\"Maps MNIST digits to a triplet (z_mean, z_log_var, z).\"\"\"\n",
"\n",
" def __init__(self, latent_dim=32, intermediate_dim=64, name=\"encoder\", **kwargs):\n",
" super(Encoder, self).__init__(name=name, **kwargs)\n",
" self.dense_proj = layers.Dense(intermediate_dim, activation=\"relu\")\n",
" self.dense_mean = layers.Dense(latent_dim)\n",
" self.dense_log_var = layers.Dense(latent_dim)\n",
" self.sampling = Sampling()\n",
"\n",
" def call(self, inputs):\n",
" x = self.dense_proj(inputs)\n",
" z_mean = self.dense_mean(x)\n",
" z_log_var = self.dense_log_var(x)\n",
" z = self.sampling((z_mean, z_log_var))\n",
" return z_mean, z_log_var, z\n",
"\n",
"\n",
"class Decoder(layers.Layer):\n",
" \"\"\"Converts z, the encoded digit vector, back into a readable digit.\"\"\"\n",
"\n",
" def __init__(self, original_dim, intermediate_dim=64, name=\"decoder\", **kwargs):\n",
" super(Decoder, self).__init__(name=name, **kwargs)\n",
" self.dense_proj = layers.Dense(intermediate_dim, activation=\"relu\")\n",
" self.dense_output = layers.Dense(original_dim, activation=\"sigmoid\")\n",
"\n",
" def call(self, inputs):\n",
" x = self.dense_proj(inputs)\n",
" return self.dense_output(x)\n",
"\n",
"\n",
"class VariationalAutoEncoder(keras.Model):\n",
" \"\"\"Combines the encoder and decoder into an end-to-end model for training.\"\"\"\n",
"\n",
" def __init__(\n",
" self,\n",
" original_dim,\n",
" intermediate_dim=64,\n",
" latent_dim=32,\n",
" name=\"autoencoder\",\n",
" **kwargs\n",
" ):\n",
" super(VariationalAutoEncoder, self).__init__(name=name, **kwargs)\n",
" self.original_dim = original_dim\n",
" self.encoder = Encoder(latent_dim=latent_dim, intermediate_dim=intermediate_dim)\n",
" self.decoder = Decoder(original_dim, intermediate_dim=intermediate_dim)\n",
"\n",
" def call(self, inputs):\n",
" z_mean, z_log_var, z = self.encoder(inputs)\n",
" reconstructed = self.decoder(z)\n",
" # Add KL divergence regularization loss.\n",
" kl_loss = -0.5 * tf.reduce_mean(\n",
" z_log_var - tf.square(z_mean) - tf.exp(z_log_var) + 1\n",
" )\n",
" self.add_loss(kl_loss)\n",
" return reconstructed\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "2f8ae035a7c9"
},
"source": [
"让我们在 MNIST 上编写一个简单的训练循环："
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:08.736274Z",
"iopub.status.busy": "2022-12-14T21:33:08.735753Z",
"iopub.status.idle": "2022-12-14T21:33:52.488196Z",
"shell.execute_reply": "2022-12-14T21:33:52.487290Z"
},
"id": "40f11d1ef3bc"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Start of epoch 0\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"WARNING:tensorflow:5 out of the last 5 calls to triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has reduce_retracing=True option that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for more details.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"WARNING:tensorflow:6 out of the last 6 calls to triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has reduce_retracing=True option that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for more details.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 0: mean loss = 0.3351\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 100: mean loss = 0.1247\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 200: mean loss = 0.0988\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 300: mean loss = 0.0890\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 400: mean loss = 0.0841\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 500: mean loss = 0.0808\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 600: mean loss = 0.0786\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 700: mean loss = 0.0771\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 800: mean loss = 0.0759\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 900: mean loss = 0.0749\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Start of epoch 1\n",
"step 0: mean loss = 0.0746\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 100: mean loss = 0.0739\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 200: mean loss = 0.0734\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 300: mean loss = 0.0730\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 400: mean loss = 0.0727\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 500: mean loss = 0.0723\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 600: mean loss = 0.0720\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 700: mean loss = 0.0717\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 800: mean loss = 0.0714\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"step 900: mean loss = 0.0712\n"
]
}
],
"source": [
"original_dim = 784\n",
"vae = VariationalAutoEncoder(original_dim, 64, 32)\n",
"\n",
"optimizer = tf.keras.optimizers.Adam(learning_rate=1e-3)\n",
"mse_loss_fn = tf.keras.losses.MeanSquaredError()\n",
"\n",
"loss_metric = tf.keras.metrics.Mean()\n",
"\n",
"(x_train, _), _ = tf.keras.datasets.mnist.load_data()\n",
"x_train = x_train.reshape(60000, 784).astype(\"float32\") / 255\n",
"\n",
"train_dataset = tf.data.Dataset.from_tensor_slices(x_train)\n",
"train_dataset = train_dataset.shuffle(buffer_size=1024).batch(64)\n",
"\n",
"epochs = 2\n",
"\n",
"# Iterate over epochs.\n",
"for epoch in range(epochs):\n",
" print(\"Start of epoch %d\" % (epoch,))\n",
"\n",
" # Iterate over the batches of the dataset.\n",
" for step, x_batch_train in enumerate(train_dataset):\n",
" with tf.GradientTape() as tape:\n",
" reconstructed = vae(x_batch_train)\n",
" # Compute reconstruction loss\n",
" loss = mse_loss_fn(x_batch_train, reconstructed)\n",
" loss += sum(vae.losses) # Add KLD regularization loss\n",
"\n",
" grads = tape.gradient(loss, vae.trainable_weights)\n",
" optimizer.apply_gradients(zip(grads, vae.trainable_weights))\n",
"\n",
" loss_metric(loss)\n",
"\n",
" if step % 100 == 0:\n",
" print(\"step %d: mean loss = %.4f\" % (step, loss_metric.result()))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "f0d65fae5d3d"
},
"source": [
"请注意，由于 VAE 是 `Model` 的子类，它具有内置的训练循环。因此，您也可以用以下方式训练它："
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:33:52.492492Z",
"iopub.status.busy": "2022-12-14T21:33:52.491934Z",
"iopub.status.idle": "2022-12-14T21:34:00.032574Z",
"shell.execute_reply": "2022-12-14T21:34:00.031850Z"
},
"id": "5af13f70d528"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/2\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\r",
" 1/938 [..............................] - ETA: 25:19 - loss: 0.3395"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 19/938 [..............................] - ETA: 2s - loss: 0.2210 "
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 38/938 [>.............................] - ETA: 2s - loss: 0.1832"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 57/938 [>.............................] - ETA: 2s - loss: 0.1579"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 76/938 [=>............................] - ETA: 2s - loss: 0.1407"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 95/938 [==>...........................] - ETA: 2s - loss: 0.1285"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"114/938 [==>...........................] - ETA: 2s - loss: 0.1197"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"133/938 [===>..........................] - ETA: 2s - loss: 0.1132"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"152/938 [===>..........................] - ETA: 2s - loss: 0.1080"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"172/938 [====>.........................] - ETA: 2s - loss: 0.1036"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"192/938 [=====>........................] - ETA: 2s - loss: 0.1002"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"211/938 [=====>........................] - ETA: 1s - loss: 0.0975"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"231/938 [======>.......................] - ETA: 1s - loss: 0.0950"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"251/938 [=======>......................] - ETA: 1s - loss: 0.0930"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"270/938 [=======>......................] - ETA: 1s - loss: 0.0912"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"289/938 [========>.....................] - ETA: 1s - loss: 0.0897"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"308/938 [========>.....................] - ETA: 1s - loss: 0.0884"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"327/938 [=========>....................] - ETA: 1s - loss: 0.0873"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"346/938 [==========>...................] - ETA: 1s - loss: 0.0862"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"366/938 [==========>...................] - ETA: 1s - loss: 0.0852"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"386/938 [===========>..................] - ETA: 1s - loss: 0.0843"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"406/938 [===========>..................] - ETA: 1s - loss: 0.0835"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"426/938 [============>.................] - ETA: 1s - loss: 0.0828"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"446/938 [=============>................] - ETA: 1s - loss: 0.0821"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"466/938 [=============>................] - ETA: 1s - loss: 0.0815"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"486/938 [==============>...............] - ETA: 1s - loss: 0.0809"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"505/938 [===============>..............] - ETA: 1s - loss: 0.0804"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"524/938 [===============>..............] - ETA: 1s - loss: 0.0800"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"543/938 [================>.............] - ETA: 1s - loss: 0.0796"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"562/938 [================>.............] - ETA: 1s - loss: 0.0792"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"581/938 [=================>............] - ETA: 0s - loss: 0.0788"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"600/938 [==================>...........] - ETA: 0s - loss: 0.0785"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"619/938 [==================>...........] - ETA: 0s - loss: 0.0782"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"638/938 [===================>..........] - ETA: 0s - loss: 0.0778"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"657/938 [====================>.........] - ETA: 0s - loss: 0.0775"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"676/938 [====================>.........] - ETA: 0s - loss: 0.0773"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"695/938 [=====================>........] - ETA: 0s - loss: 0.0770"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"714/938 [=====================>........] - ETA: 0s - loss: 0.0768"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"732/938 [======================>.......] - ETA: 0s - loss: 0.0765"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"751/938 [=======================>......] - ETA: 0s - loss: 0.0763"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"770/938 [=======================>......] - ETA: 0s - loss: 0.0761"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"789/938 [========================>.....] - ETA: 0s - loss: 0.0759"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"808/938 [========================>.....] - ETA: 0s - loss: 0.0758"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"827/938 [=========================>....] - ETA: 0s - loss: 0.0756"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"846/938 [==========================>...] - ETA: 0s - loss: 0.0754"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"865/938 [==========================>...] - ETA: 0s - loss: 0.0753"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"884/938 [===========================>..] - ETA: 0s - loss: 0.0751"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"903/938 [===========================>..] - ETA: 0s - loss: 0.0750"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"922/938 [============================>.] - ETA: 0s - loss: 0.0749"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"938/938 [==============================] - 4s 3ms/step - loss: 0.0747\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 2/2\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\r",
" 1/938 [..............................] - ETA: 3s - loss: 0.0669"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 21/938 [..............................] - ETA: 2s - loss: 0.0681"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 40/938 [>.............................] - ETA: 2s - loss: 0.0679"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 59/938 [>.............................] - ETA: 2s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 78/938 [=>............................] - ETA: 2s - loss: 0.0674"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 96/938 [==>...........................] - ETA: 2s - loss: 0.0674"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"115/938 [==>...........................] - ETA: 2s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"133/938 [===>..........................] - ETA: 2s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"151/938 [===>..........................] - ETA: 2s - loss: 0.0674"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"170/938 [====>.........................] - ETA: 2s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"188/938 [=====>........................] - ETA: 2s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"207/938 [=====>........................] - ETA: 2s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"226/938 [======>.......................] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"245/938 [======>.......................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"263/938 [=======>......................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"282/938 [========>.....................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"300/938 [========>.....................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"319/938 [=========>....................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"338/938 [=========>....................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"356/938 [==========>...................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"374/938 [==========>...................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"392/938 [===========>..................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"410/938 [============>.................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"428/938 [============>.................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"446/938 [=============>................] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"464/938 [=============>................] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"482/938 [==============>...............] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"500/938 [==============>...............] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"518/938 [===============>..............] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"537/938 [================>.............] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"555/938 [================>.............] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"573/938 [=================>............] - ETA: 1s - loss: 0.0677"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"592/938 [=================>............] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"611/938 [==================>...........] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"629/938 [===================>..........] - ETA: 0s - loss: 0.0677"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"648/938 [===================>..........] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"667/938 [====================>.........] - ETA: 0s - loss: 0.0677"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"686/938 [====================>.........] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"705/938 [=====================>........] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"723/938 [======================>.......] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"741/938 [======================>.......] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"759/938 [=======================>......] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"778/938 [=======================>......] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"797/938 [========================>.....] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"815/938 [=========================>....] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"834/938 [=========================>....] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"853/938 [==========================>...] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"872/938 [==========================>...] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"891/938 [===========================>..] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"909/938 [============================>.] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"928/938 [============================>.] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"938/938 [==============================] - 3s 3ms/step - loss: 0.0676\n"
]
},
{
"data": {
"text/plain": [
""
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"vae = VariationalAutoEncoder(784, 64, 32)\n",
"\n",
"optimizer = tf.keras.optimizers.Adam(learning_rate=1e-3)\n",
"\n",
"vae.compile(optimizer, loss=tf.keras.losses.MeanSquaredError())\n",
"vae.fit(x_train, x_train, epochs=2, batch_size=64)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "d34b7ba21662"
},
"source": [
"## 超越面向对象的开发：函数式 API\n",
"\n",
"这个示例对您来说是否包含了太多面向对象的开发？您也可以使用[函数式 API](https://tensorflow.google.cn/guide/keras/functional/) 来构建模型。重要的是，选择其中一种样式并不妨碍您利用以另一种样式编写的组件：您随时可以搭配使用。\n",
"\n",
"例如，下面的函数式 API 示例重用了我们在上面的示例中定义的同一个 `Sampling` 层："
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-14T21:34:00.036230Z",
"iopub.status.busy": "2022-12-14T21:34:00.035692Z",
"iopub.status.idle": "2022-12-14T21:34:10.244554Z",
"shell.execute_reply": "2022-12-14T21:34:10.243702Z"
},
"id": "be77fc8f9b26"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/3\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\r",
" 1/938 [..............................] - ETA: 23:18 - loss: 0.3739"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 19/938 [..............................] - ETA: 2s - loss: 0.2266 "
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 37/938 [>.............................] - ETA: 2s - loss: 0.1877"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 55/938 [>.............................] - ETA: 2s - loss: 0.1622"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 74/938 [=>............................] - ETA: 2s - loss: 0.1436"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 93/938 [=>............................] - ETA: 2s - loss: 0.1309"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"112/938 [==>...........................] - ETA: 2s - loss: 0.1217"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"131/938 [===>..........................] - ETA: 2s - loss: 0.1147"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"150/938 [===>..........................] - ETA: 2s - loss: 0.1094"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"169/938 [====>.........................] - ETA: 2s - loss: 0.1052"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"188/938 [=====>........................] - ETA: 2s - loss: 0.1016"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"207/938 [=====>........................] - ETA: 2s - loss: 0.0986"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"226/938 [======>.......................] - ETA: 1s - loss: 0.0962"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"245/938 [======>.......................] - ETA: 1s - loss: 0.0940"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"264/938 [=======>......................] - ETA: 1s - loss: 0.0923"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"282/938 [========>.....................] - ETA: 1s - loss: 0.0908"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"301/938 [========>.....................] - ETA: 1s - loss: 0.0895"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"320/938 [=========>....................] - ETA: 1s - loss: 0.0882"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"339/938 [=========>....................] - ETA: 1s - loss: 0.0871"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"358/938 [==========>...................] - ETA: 1s - loss: 0.0861"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"377/938 [===========>..................] - ETA: 1s - loss: 0.0852"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"396/938 [===========>..................] - ETA: 1s - loss: 0.0844"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"414/938 [============>.................] - ETA: 1s - loss: 0.0837"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"433/938 [============>.................] - ETA: 1s - loss: 0.0830"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"452/938 [=============>................] - ETA: 1s - loss: 0.0823"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"471/938 [==============>...............] - ETA: 1s - loss: 0.0817"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"490/938 [==============>...............] - ETA: 1s - loss: 0.0812"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"509/938 [===============>..............] - ETA: 1s - loss: 0.0807"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"528/938 [===============>..............] - ETA: 1s - loss: 0.0803"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"546/938 [================>.............] - ETA: 1s - loss: 0.0798"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"564/938 [=================>............] - ETA: 1s - loss: 0.0794"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"582/938 [=================>............] - ETA: 0s - loss: 0.0790"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"601/938 [==================>...........] - ETA: 0s - loss: 0.0787"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"620/938 [==================>...........] - ETA: 0s - loss: 0.0784"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"638/938 [===================>..........] - ETA: 0s - loss: 0.0780"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"657/938 [====================>.........] - ETA: 0s - loss: 0.0777"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"675/938 [====================>.........] - ETA: 0s - loss: 0.0775"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"694/938 [=====================>........] - ETA: 0s - loss: 0.0772"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"713/938 [=====================>........] - ETA: 0s - loss: 0.0770"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"732/938 [======================>.......] - ETA: 0s - loss: 0.0768"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"750/938 [======================>.......] - ETA: 0s - loss: 0.0765"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"768/938 [=======================>......] - ETA: 0s - loss: 0.0763"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"786/938 [========================>.....] - ETA: 0s - loss: 0.0761"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"804/938 [========================>.....] - ETA: 0s - loss: 0.0760"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"822/938 [=========================>....] - ETA: 0s - loss: 0.0758"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"840/938 [=========================>....] - ETA: 0s - loss: 0.0756"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"859/938 [==========================>...] - ETA: 0s - loss: 0.0754"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"877/938 [===========================>..] - ETA: 0s - loss: 0.0753"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"894/938 [===========================>..] - ETA: 0s - loss: 0.0751"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"912/938 [============================>.] - ETA: 0s - loss: 0.0750"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"931/938 [============================>.] - ETA: 0s - loss: 0.0748"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"938/938 [==============================] - 4s 3ms/step - loss: 0.0748\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 2/3\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\r",
" 1/938 [..............................] - ETA: 3s - loss: 0.0650"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 20/938 [..............................] - ETA: 2s - loss: 0.0673"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 38/938 [>.............................] - ETA: 2s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 57/938 [>.............................] - ETA: 2s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 76/938 [=>............................] - ETA: 2s - loss: 0.0677"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 94/938 [==>...........................] - ETA: 2s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"112/938 [==>...........................] - ETA: 2s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"131/938 [===>..........................] - ETA: 2s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"150/938 [===>..........................] - ETA: 2s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"168/938 [====>.........................] - ETA: 2s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"186/938 [====>.........................] - ETA: 2s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"204/938 [=====>........................] - ETA: 2s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"222/938 [======>.......................] - ETA: 2s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"240/938 [======>.......................] - ETA: 1s - loss: 0.0679"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"258/938 [=======>......................] - ETA: 1s - loss: 0.0679"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"276/938 [=======>......................] - ETA: 1s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"295/938 [========>.....................] - ETA: 1s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"313/938 [=========>....................] - ETA: 1s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"331/938 [=========>....................] - ETA: 1s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"350/938 [==========>...................] - ETA: 1s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"368/938 [==========>...................] - ETA: 1s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"386/938 [===========>..................] - ETA: 1s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"404/938 [===========>..................] - ETA: 1s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"422/938 [============>.................] - ETA: 1s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"441/938 [=============>................] - ETA: 1s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"459/938 [=============>................] - ETA: 1s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"477/938 [==============>...............] - ETA: 1s - loss: 0.0678"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"496/938 [==============>...............] - ETA: 1s - loss: 0.0677"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"514/938 [===============>..............] - ETA: 1s - loss: 0.0677"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"532/938 [================>.............] - ETA: 1s - loss: 0.0677"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"550/938 [================>.............] - ETA: 1s - loss: 0.0677"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"568/938 [=================>............] - ETA: 1s - loss: 0.0677"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"586/938 [=================>............] - ETA: 0s - loss: 0.0677"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"604/938 [==================>...........] - ETA: 0s - loss: 0.0677"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"622/938 [==================>...........] - ETA: 0s - loss: 0.0677"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"640/938 [===================>..........] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"658/938 [====================>.........] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"676/938 [====================>.........] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"694/938 [=====================>........] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"712/938 [=====================>........] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"731/938 [======================>.......] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"749/938 [======================>.......] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"767/938 [=======================>......] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"785/938 [========================>.....] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"803/938 [========================>.....] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"821/938 [=========================>....] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"839/938 [=========================>....] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"857/938 [==========================>...] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"875/938 [==========================>...] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"893/938 [===========================>..] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"911/938 [============================>.] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"929/938 [============================>.] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"938/938 [==============================] - 3s 3ms/step - loss: 0.0676\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 3/3\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\r",
" 1/938 [..............................] - ETA: 3s - loss: 0.0672"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 19/938 [..............................] - ETA: 2s - loss: 0.0681"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 37/938 [>.............................] - ETA: 2s - loss: 0.0677"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 55/938 [>.............................] - ETA: 2s - loss: 0.0673"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 74/938 [=>............................] - ETA: 2s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 93/938 [=>............................] - ETA: 2s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"111/938 [==>...........................] - ETA: 2s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"129/938 [===>..........................] - ETA: 2s - loss: 0.0677"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"148/938 [===>..........................] - ETA: 2s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"167/938 [====>.........................] - ETA: 2s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"185/938 [====>.........................] - ETA: 2s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"203/938 [=====>........................] - ETA: 2s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"221/938 [======>.......................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"239/938 [======>.......................] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"257/938 [=======>......................] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"275/938 [=======>......................] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"294/938 [========>.....................] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"312/938 [========>.....................] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"331/938 [=========>....................] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"349/938 [==========>...................] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"367/938 [==========>...................] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"386/938 [===========>..................] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"404/938 [===========>..................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"422/938 [============>.................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"441/938 [=============>................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"460/938 [=============>................] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"479/938 [==============>...............] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"498/938 [==============>...............] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"516/938 [===============>..............] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"534/938 [================>.............] - ETA: 1s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"552/938 [================>.............] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"570/938 [=================>............] - ETA: 1s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"588/938 [=================>............] - ETA: 0s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"606/938 [==================>...........] - ETA: 0s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"625/938 [==================>...........] - ETA: 0s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"643/938 [===================>..........] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"662/938 [====================>.........] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"681/938 [====================>.........] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"699/938 [=====================>........] - ETA: 0s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"717/938 [=====================>........] - ETA: 0s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"735/938 [======================>.......] - ETA: 0s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"753/938 [=======================>......] - ETA: 0s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"771/938 [=======================>......] - ETA: 0s - loss: 0.0675"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"789/938 [========================>.....] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"808/938 [========================>.....] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"827/938 [=========================>....] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"846/938 [==========================>...] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"865/938 [==========================>...] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"884/938 [===========================>..] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"903/938 [===========================>..] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"922/938 [============================>.] - ETA: 0s - loss: 0.0676"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"938/938 [==============================] - 3s 3ms/step - loss: 0.0676\n"
]
},
{
"data": {
"text/plain": [
""
]
},
"execution_count": 26,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"original_dim = 784\n",
"intermediate_dim = 64\n",
"latent_dim = 32\n",
"\n",
"# Define encoder model.\n",
"original_inputs = tf.keras.Input(shape=(original_dim,), name=\"encoder_input\")\n",
"x = layers.Dense(intermediate_dim, activation=\"relu\")(original_inputs)\n",
"z_mean = layers.Dense(latent_dim, name=\"z_mean\")(x)\n",
"z_log_var = layers.Dense(latent_dim, name=\"z_log_var\")(x)\n",
"z = Sampling()((z_mean, z_log_var))\n",
"encoder = tf.keras.Model(inputs=original_inputs, outputs=z, name=\"encoder\")\n",
"\n",
"# Define decoder model.\n",
"latent_inputs = tf.keras.Input(shape=(latent_dim,), name=\"z_sampling\")\n",
"x = layers.Dense(intermediate_dim, activation=\"relu\")(latent_inputs)\n",
"outputs = layers.Dense(original_dim, activation=\"sigmoid\")(x)\n",
"decoder = tf.keras.Model(inputs=latent_inputs, outputs=outputs, name=\"decoder\")\n",
"\n",
"# Define VAE model.\n",
"outputs = decoder(z)\n",
"vae = tf.keras.Model(inputs=original_inputs, outputs=outputs, name=\"vae\")\n",
"\n",
"# Add KL divergence regularization loss.\n",
"kl_loss = -0.5 * tf.reduce_mean(z_log_var - tf.square(z_mean) - tf.exp(z_log_var) + 1)\n",
"vae.add_loss(kl_loss)\n",
"\n",
"# Train.\n",
"optimizer = tf.keras.optimizers.Adam(learning_rate=1e-3)\n",
"vae.compile(optimizer, loss=tf.keras.losses.MeanSquaredError())\n",
"vae.fit(x_train, x_train, epochs=3, batch_size=64)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "e2f135ea7cf5"
},
"source": [
"有关详情，请务必阅读[函数式 API](https://tensorflow.google.cn/guide/keras/functional/) 指南。"
]
}
],
"metadata": {
"colab": {
"collapsed_sections": [],
"name": "custom_layers_and_models.ipynb",
"toc_visible": true
},
"kernelspec": {
"display_name": "Python 3",
"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": "3.9.16"
}
},
"nbformat": 4,
"nbformat_minor": 0
}
![](\"https://tensorflow.google.cn/images/tf_logo_32px.png\"){kind=logo}
在 TensorFlow.org 上查看 ![](\"https://tensorflow.google.cn/images/colab_logo_32px.png\"){kind=logo}
在 Google Colab 中运行 ![](\"https://tensorflow.google.cn/images/GitHub-Mark-32px.png\")
在 GitHub 上查看源代码 ![](\"https://tensorflow.google.cn/images/download_logo_32px.png\"){kind=logo}
下载笔记本