<br />

|---|---|---|---|
| [![](https://www.tensorflow.org/images/tf_logo_32px.png)View on TensorFlow.org](https://www.tensorflow.org/tutorials/load_data/numpy) | [![](https://www.tensorflow.org/images/colab_logo_32px.png)Run in Google Colab](https://colab.research.google.com/github/tensorflow/docs/blob/master/site/en/tutorials/load_data/numpy.ipynb) | [![](https://www.tensorflow.org/images/GitHub-Mark-32px.png)View source on GitHub](https://github.com/tensorflow/docs/blob/master/site/en/tutorials/load_data/numpy.ipynb) | [![](https://www.tensorflow.org/images/download_logo_32px.png)Download notebook](https://storage.googleapis.com/tensorflow_docs/docs/site/en/tutorials/load_data/numpy.ipynb) |

This tutorial provides an example of loading data from NumPy arrays into a [`tf.data.Dataset`](https://www.tensorflow.org/api_docs/python/tf/data/Dataset).

This example loads the MNIST dataset from a `.npz` file. However, the source of the NumPy arrays is not important.

## Setup

    import numpy as np
    import tensorflow as tf

```
2025-12-01 08:37:36.241605: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:467] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
E0000 00:00:1764578256.266252  382952 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
E0000 00:00:1764578256.273818  382952 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
W0000 00:00:1764578256.292643  382952 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1764578256.292669  382952 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1764578256.292671  382952 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1764578256.292674  382952 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
```

### Load from `.npz` file

    DATA_URL = 'https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz'

    path = tf.keras.utils.get_file('mnist.npz', DATA_URL)
    with np.load(path) as data:
      train_examples = data['x_train']
      train_labels = data['y_train']
      test_examples = data['x_test']
      test_labels = data['y_test']

## Load NumPy arrays with [`tf.data.Dataset`](https://www.tensorflow.org/api_docs/python/tf/data/Dataset)

Assuming you have an array of examples and a corresponding array of labels, pass the two arrays as a tuple into [`tf.data.Dataset.from_tensor_slices`](https://www.tensorflow.org/api_docs/python/tf/data/Dataset#from_tensor_slices) to create a [`tf.data.Dataset`](https://www.tensorflow.org/api_docs/python/tf/data/Dataset).

    train_dataset = tf.data.Dataset.from_tensor_slices((train_examples, train_labels))
    test_dataset = tf.data.Dataset.from_tensor_slices((test_examples, test_labels))

```
I0000 00:00:1764578260.988159  382952 gpu_device.cc:2019] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 13638 MB memory:  -> device: 0, name: Tesla T4, pci bus id: 0000:00:05.0, compute capability: 7.5
I0000 00:00:1764578260.990477  382952 gpu_device.cc:2019] Created device /job:localhost/replica:0/task:0/device:GPU:1 with 13756 MB memory:  -> device: 1, name: Tesla T4, pci bus id: 0000:00:06.0, compute capability: 7.5
I0000 00:00:1764578260.992652  382952 gpu_device.cc:2019] Created device /job:localhost/replica:0/task:0/device:GPU:2 with 13756 MB memory:  -> device: 2, name: Tesla T4, pci bus id: 0000:00:07.0, compute capability: 7.5
I0000 00:00:1764578260.995006  382952 gpu_device.cc:2019] Created device /job:localhost/replica:0/task:0/device:GPU:3 with 13756 MB memory:  -> device: 3, name: Tesla T4, pci bus id: 0000:00:08.0, compute capability: 7.5
```

## Use the datasets

### Shuffle and batch the datasets

    BATCH_SIZE = 64
    SHUFFLE_BUFFER_SIZE = 100

    train_dataset = train_dataset.shuffle(SHUFFLE_BUFFER_SIZE).batch(BATCH_SIZE)
    test_dataset = test_dataset.batch(BATCH_SIZE)

### Build and train a model

    model = tf.keras.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.RMSprop(),
                  loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
                  metrics=['sparse_categorical_accuracy'])

```
/tmpfs/src/tf_docs_env/lib/python3.9/site-packages/keras/src/layers/reshaping/flatten.py:37: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.
  super().__init__(**kwargs)
```

    model.fit(train_dataset, epochs=10)

```
Epoch 1/10
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
I0000 00:00:1764578262.725289  383118 service.cc:152] XLA service 0x7efa64005330 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:
I0000 00:00:1764578262.725320  383118 service.cc:160]   StreamExecutor device (0): Tesla T4, Compute Capability 7.5
I0000 00:00:1764578262.725324  383118 service.cc:160]   StreamExecutor device (1): Tesla T4, Compute Capability 7.5
I0000 00:00:1764578262.725326  383118 service.cc:160]   StreamExecutor device (2): Tesla T4, Compute Capability 7.5
I0000 00:00:1764578262.725335  383118 service.cc:160]   StreamExecutor device (3): Tesla T4, Compute Capability 7.5
I0000 00:00:1764578262.810544  383118 cuda_dnn.cc:529] Loaded cuDNN version 91600
80/938 ━━━━━━━━━━━━━━━━━━━━ 1s 2ms/step - loss: 33.2599 - sparse_categorical_accuracy: 0.6180
I0000 00:00:1764578263.427021  383118 device_compiler.h:188] Compiled cluster using XLA!  This line is logged at most once for the lifetime of the process.
938/938 ━━━━━━━━━━━━━━━━━━━━ 4s 3ms/step - loss: 8.3610 - sparse_categorical_accuracy: 0.8231
Epoch 2/10
938/938 ━━━━━━━━━━━━━━━━━━━━ 2s 2ms/step - loss: 0.5359 - sparse_categorical_accuracy: 0.9176
Epoch 3/10
938/938 ━━━━━━━━━━━━━━━━━━━━ 2s 2ms/step - loss: 0.3971 - sparse_categorical_accuracy: 0.9380
Epoch 4/10
938/938 ━━━━━━━━━━━━━━━━━━━━ 2s 2ms/step - loss: 0.3160 - sparse_categorical_accuracy: 0.9518
Epoch 5/10
938/938 ━━━━━━━━━━━━━━━━━━━━ 2s 2ms/step - loss: 0.2804 - sparse_categorical_accuracy: 0.9571
Epoch 6/10
938/938 ━━━━━━━━━━━━━━━━━━━━ 2s 2ms/step - loss: 0.2402 - sparse_categorical_accuracy: 0.9613
Epoch 7/10
938/938 ━━━━━━━━━━━━━━━━━━━━ 2s 2ms/step - loss: 0.2256 - sparse_categorical_accuracy: 0.9659
Epoch 8/10
938/938 ━━━━━━━━━━━━━━━━━━━━ 2s 2ms/step - loss: 0.2261 - sparse_categorical_accuracy: 0.9674
Epoch 9/10
938/938 ━━━━━━━━━━━━━━━━━━━━ 2s 2ms/step - loss: 0.2035 - sparse_categorical_accuracy: 0.9708
Epoch 10/10
938/938 ━━━━━━━━━━━━━━━━━━━━ 2s 2ms/step - loss: 0.1763 - sparse_categorical_accuracy: 0.9735
<keras.src.callbacks.history.History at 0x7efd6809a1c0>
```

    model.evaluate(test_dataset)

```
157/157 ━━━━━━━━━━━━━━━━━━━━ 1s 5ms/step - loss: 0.5675 - sparse_categorical_accuracy: 0.9495
[0.527810275554657, 0.953499972820282]
```