TFDS now supports the Croissant 🥐 format! Read the documentation to know more.
tfds.as_numpy
Stay organized with collections
Save and categorize content based on your preferences.
Converts a tf.data.Dataset to an iterable of NumPy arrays.
tfds.as_numpy(
dataset: Tree[TensorflowElem]
) -> Tree[NumpyElem]
Used in the notebooks
| Used in the tutorials |
|---|
as_numpy converts a possibly nested structure of tf.data.Datasets
and tf.Tensors to iterables of NumPy arrays and NumPy arrays, respectively.
Note that because TensorFlow has support for ragged tensors and NumPy has
no equivalent representation,
tf.RaggedTensors
are left as-is for the user to deal with them (e.g. using to_list()).
In TF 1 (i.e. graph mode), tf.RaggedTensors are returned as
tf.ragged.RaggedTensorValues.
Example:
ds = tfds.load(name="mnist", split="train")
ds_numpy = tfds.as_numpy(ds) # Convert `tf.data.Dataset` to Python generator
for ex in ds_numpy:
# `{'image': np.array(shape=(28, 28, 1)), 'labels': np.array(shape=())}`
print(ex)
Args | |
|---|---|
dataset
|
a possibly nested structure of tf.data.Datasets and/or
tf.Tensors.
|
Returns | |
|---|---|
A structure matching dataset where tf.data.Datasets are converted to
generators of NumPy arrays and tf.Tensors are converted to NumPy arrays.
|