Spaces:
Sleeping
Sleeping
File size: 8,168 Bytes
4d4d14c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 |
{
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "USSV_OlCFKOD"
},
"source": [
"# Training a neural network on MNIST with Keras\n",
"\n",
"This simple example demonstrates how to plug TensorFlow Datasets (TFDS) into a Keras model.\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "J8y9ZkLXmAZc"
},
"source": [
"Copyright 2020 The TensorFlow Datasets Authors, Licensed under the Apache License, Version 2.0"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "OGw9EgE0tC0C"
},
"source": [
"<table class=\"tfo-notebook-buttons\" align=\"left\">\n",
" <td>\n",
" <a target=\"_blank\" href=\"https://www.tensorflow.org/datasets/keras_example\"><img src=\"https://www.tensorflow.org/images/tf_logo_32px.png\" />View on TensorFlow.org</a>\n",
" </td>\n",
" <td>\n",
" <a target=\"_blank\" href=\"https://colab.research.google.com/github/tensorflow/datasets/blob/master/docs/keras_example.ipynb\"><img src=\"https://www.tensorflow.org/images/colab_logo_32px.png\" />Run in Google Colab</a>\n",
" </td>\n",
" <td>\n",
" <a target=\"_blank\" href=\"https://github.com/tensorflow/datasets/blob/master/docs/keras_example.ipynb\"><img src=\"https://www.tensorflow.org/images/GitHub-Mark-32px.png\" />View source on GitHub</a>\n",
" </td>\n",
" <td>\n",
" <a href=\"https://storage.googleapis.com/tensorflow_docs/datasets/docs/keras_example.ipynb\"><img src=\"https://www.tensorflow.org/images/download_logo_32px.png\" />Download notebook</a>\n",
" </td>\n",
"</table>"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "TTBSvHcSLBzc"
},
"outputs": [],
"source": [
"import tensorflow as tf\n",
"import tensorflow_datasets as tfds"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "VjI6VgOBf0v0"
},
"source": [
"## Step 1: Create your input pipeline\n",
"\n",
"Start by building an efficient input pipeline using advices from:\n",
"* The [Performance tips](https://www.tensorflow.org/datasets/performances) guide\n",
"* The [Better performance with the `tf.data` API](https://www.tensorflow.org/guide/data_performance#optimize_performance) guide\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "c3aH3vP_XLI8"
},
"source": [
"### Load a dataset\n",
"\n",
"Load the MNIST dataset with the following arguments:\n",
"\n",
"* `shuffle_files=True`: The MNIST data is only stored in a single file, but for larger datasets with multiple files on disk, it's good practice to shuffle them when training.\n",
"* `as_supervised=True`: Returns a tuple `(img, label)` instead of a dictionary `{'image': img, 'label': label}`."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "ZUMhCXhFXdHQ"
},
"outputs": [],
"source": [
"(ds_train, ds_test), ds_info = tfds.load(\n",
" 'mnist',\n",
" split=['train', 'test'],\n",
" shuffle_files=True,\n",
" as_supervised=True,\n",
" with_info=True,\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "rgwCFAcWXQTx"
},
"source": [
"### Build a training pipeline\n",
"\n",
"Apply the following transformations:\n",
"\n",
"* `tf.data.Dataset.map`: TFDS provide images of type `tf.uint8`, while the model expects `tf.float32`. Therefore, you need to normalize images.\n",
"* `tf.data.Dataset.cache` As you fit the dataset in memory, cache it before shuffling for a better performance.<br/>\n",
"__Note:__ Random transformations should be applied after caching.\n",
"* `tf.data.Dataset.shuffle`: For true randomness, set the shuffle buffer to the full dataset size.<br/>\n",
"__Note:__ For large datasets that can't fit in memory, use `buffer_size=1000` if your system allows it.\n",
"* `tf.data.Dataset.batch`: Batch elements of the dataset after shuffling to get unique batches at each epoch.\n",
"* `tf.data.Dataset.prefetch`: It is good practice to end the pipeline by prefetching [for performance](https://www.tensorflow.org/guide/data_performance#prefetching)."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "haykx2K9XgiI"
},
"outputs": [],
"source": [
"def normalize_img(image, label):\n",
" \"\"\"Normalizes images: `uint8` -> `float32`.\"\"\"\n",
" return tf.cast(image, tf.float32) / 255., label\n",
"\n",
"ds_train = ds_train.map(\n",
" normalize_img, num_parallel_calls=tf.data.AUTOTUNE)\n",
"ds_train = ds_train.cache()\n",
"ds_train = ds_train.shuffle(ds_info.splits['train'].num_examples)\n",
"ds_train = ds_train.batch(128)\n",
"ds_train = ds_train.prefetch(tf.data.AUTOTUNE)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "RbsMy4X1XVFv"
},
"source": [
"### Build an evaluation pipeline\n",
"\n",
"Your testing pipeline is similar to the training pipeline with small differences:\n",
"\n",
" * You don't need to call `tf.data.Dataset.shuffle`.\n",
" * Caching is done after batching because batches can be the same between epochs."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "A0KjuDf7XiqY"
},
"outputs": [],
"source": [
"ds_test = ds_test.map(\n",
" normalize_img, num_parallel_calls=tf.data.AUTOTUNE)\n",
"ds_test = ds_test.batch(128)\n",
"ds_test = ds_test.cache()\n",
"ds_test = ds_test.prefetch(tf.data.AUTOTUNE)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "nTFoji3INMEM"
},
"source": [
"## Step 2: Create and train the model\n",
"\n",
"Plug the TFDS input pipeline into a simple Keras model, compile the model, and train it."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "XWqxdmS1NLKA"
},
"outputs": [],
"source": [
"model = tf.keras.models.Sequential([\n",
" tf.keras.layers.Flatten(input_shape=(28, 28)),\n",
" tf.keras.layers.Dense(128, activation='relu'),\n",
" tf.keras.layers.Dense(10)\n",
"])\n",
"model.compile(\n",
" optimizer=tf.keras.optimizers.Adam(0.001),\n",
" loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
" metrics=[tf.keras.metrics.SparseCategoricalAccuracy()],\n",
")\n",
"\n",
"model.fit(\n",
" ds_train,\n",
" epochs=6,\n",
" validation_data=ds_test,\n",
")"
]
},
{
"cell_type": "markdown",
"source": [
"Save Model Weights"
],
"metadata": {
"id": "lOLnm8sk-rDP"
}
},
{
"cell_type": "code",
"source": [
"# save model\n",
"model.save('model.h5')"
],
"metadata": {
"id": "8nd9iSyG-s9p"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [],
"metadata": {
"id": "DQuEn_g7-vlR"
},
"execution_count": null,
"outputs": []
}
],
"metadata": {
"colab": {
"private_outputs": true,
"provenance": [],
"toc_visible": true
},
"kernelspec": {
"display_name": "Python 3",
"name": "python3"
}
},
"nbformat": 4,
"nbformat_minor": 0
} |