Training a neural network on MNIST with Keras

This simple example demonstrates how to plug TensorFlow Datasets (TFDS) into a Keras model.

View on TensorFlow.org

Run in Google Colab

View source on GitHub

Download notebook

import tensorflow as tf
import tensorflow_datasets as tfds

Step 1: Create your input pipeline

Start by building an efficient input pipeline using advices from:

• The Performance tips guide
• The Better performance with the tf.data API guide

Load a dataset

Load the MNIST dataset with the following arguments:

• 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.
• as_supervised=True: Returns a tuple (img, label) instead of a dictionary {'image': img, 'label': label}.

(ds_train, ds_test), ds_info = tfds.load(
    'mnist',
    split=['train', 'test'],
    shuffle_files=True,
    as_supervised=True,
    with_info=True,
)

2023-04-14 11:14:48.760452: E tensorflow/compiler/xla/stream_executor/cuda/cuda_driver.cc:266] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected

Build a training pipeline

Apply the following transformations:

• tf.data.Dataset.map: TFDS provide images of type tf.uint8, while the model expects tf.float32. Therefore, you need to normalize images.
• tf.data.Dataset.cache As you fit the dataset in memory, cache it before shuffling for a better performance.
  Note: Random transformations should be applied after caching.
• tf.data.Dataset.shuffle: For true randomness, set the shuffle buffer to the full dataset size.
  Note: For large datasets that can't fit in memory, use buffer_size=1000 if your system allows it.
• tf.data.Dataset.batch: Batch elements of the dataset after shuffling to get unique batches at each epoch.
• tf.data.Dataset.prefetch: It is good practice to end the pipeline by prefetching for performance.

def normalize_img(image, label):
  """Normalizes images: `uint8` -> `float32`."""
  return tf.cast(image, tf.float32) / 255., label

ds_train = ds_train.map(
    normalize_img, num_parallel_calls=tf.data.AUTOTUNE)
ds_train = ds_train.cache()
ds_train = ds_train.shuffle(ds_info.splits['train'].num_examples)
ds_train = ds_train.batch(128)
ds_train = ds_train.prefetch(tf.data.AUTOTUNE)

Build an evaluation pipeline

Your testing pipeline is similar to the training pipeline with small differences:

• You don't need to call tf.data.Dataset.shuffle.
• Caching is done after batching because batches can be the same between epochs.

ds_test = ds_test.map(
    normalize_img, num_parallel_calls=tf.data.AUTOTUNE)
ds_test = ds_test.batch(128)
ds_test = ds_test.cache()
ds_test = ds_test.prefetch(tf.data.AUTOTUNE)

Step 2: Create and train the model

Plug the TFDS input pipeline into a simple Keras model, compile the model, and train it.

model = tf.keras.models.Sequential([
  tf.keras.layers.Flatten(input_shape=(28, 28)),
  tf.keras.layers.Dense(128, activation='relu'),
  tf.keras.layers.Dense(10)
])
model.compile(
    optimizer=tf.keras.optimizers.Adam(0.001),
    loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
    metrics=[tf.keras.metrics.SparseCategoricalAccuracy()],
)

model.fit(
    ds_train,
    epochs=6,
    validation_data=ds_test,
)

Epoch 1/6
469/469 [==============================] - 5s 4ms/step - loss: 0.3545 - sparse_categorical_accuracy: 0.9016 - val_loss: 0.1886 - val_sparse_categorical_accuracy: 0.9458
Epoch 2/6
469/469 [==============================] - 1s 2ms/step - loss: 0.1593 - sparse_categorical_accuracy: 0.9536 - val_loss: 0.1307 - val_sparse_categorical_accuracy: 0.9605
Epoch 3/6
469/469 [==============================] - 1s 2ms/step - loss: 0.1141 - sparse_categorical_accuracy: 0.9668 - val_loss: 0.1100 - val_sparse_categorical_accuracy: 0.9675
Epoch 4/6
469/469 [==============================] - 1s 2ms/step - loss: 0.0890 - sparse_categorical_accuracy: 0.9742 - val_loss: 0.0925 - val_sparse_categorical_accuracy: 0.9729
Epoch 5/6
469/469 [==============================] - 1s 2ms/step - loss: 0.0729 - sparse_categorical_accuracy: 0.9784 - val_loss: 0.0825 - val_sparse_categorical_accuracy: 0.9753
Epoch 6/6
469/469 [==============================] - 1s 2ms/step - loss: 0.0612 - sparse_categorical_accuracy: 0.9818 - val_loss: 0.0797 - val_sparse_categorical_accuracy: 0.9751
<keras.callbacks.History at 0x7f7aa9801ac0>