Represents a potentially large set of elements.

Used in the notebooks

Used in the guide Used in the tutorials

The API supports writing descriptive and efficient input pipelines. Dataset usage follows a common pattern:

  1. Create a source dataset from your input data.
  2. Apply dataset transformations to preprocess the data.
  3. Iterate over the dataset and process the elements.

Iteration happens in a streaming fashion, so the full dataset does not need to fit into memory.

Source Datasets:

The simplest way to create a dataset is to create it from a python list:

dataset =[1, 2, 3])
for element in dataset:
tf.Tensor(1, shape=(), dtype=int32)
tf.Tensor(2, shape=(), dtype=int32)
tf.Tensor(3, shape=(), dtype=int32)

To process lines from files, use

dataset =["file1.txt", "file2.txt"])

To process records written in the TFRecord format, use TFRecordDataset:

dataset =["file1.tfrecords", "file2.tfrecords"])

To create a dataset of all files matching a pattern, use

dataset ="/path/*.txt")  # doctest: +SKIP

See and for more ways to create datasets.


Once you have a dataset, you can apply transformations to prepare the data for your model:

dataset =[1, 2, 3])
dataset = x: x*2)
[2, 4, 6]

Common Terms:

Element: A single output from calling next() on a dataset iterator. Elements may be nested structures containing multiple components. For example, the element (1, (3, "apple")) has one tuple nested in another tuple. The components are 1, 3, and "apple".

Component: The leaf in the nested structure of an element.

Supported types:

Elements can be nested structures of tuples, named tuples, and dictionaries. Note that Python lists are not treated as nested structures of components. Instead, lists are converted to tensors and treated as components. For example, the element (1, [1, 2, 3]) has only two components; the tensor