Input/Output

Synthetic Data

range	Creates a Dataset from a range of integers [0..n).
range_tensor	Creates a Dataset tensors of the provided shape from range [0...n].

Python Objects

from_items

Create a Dataset from a list of local Python objects.

Parquet

read_parquet	Creates a Dataset from parquet files.
read_parquet_bulk	Create Dataset from parquet files without reading metadata.
Dataset.write_parquet	Writes the Dataset to parquet files under the provided path.

CSV

read_csv	Creates a Dataset from CSV files.
Dataset.write_csv	Writes the Dataset to CSV files.

JSON

read_json	Creates a Dataset from JSON and JSONL files.
Dataset.write_json	Writes the Dataset to JSON and JSONL files.

Text

read_text

Create a Dataset from lines stored in text files.

Avro

read_avro

Create a Dataset from records stored in Avro files.

Images

read_images	Creates a Dataset from image files.
Dataset.write_images	Writes the Dataset to images.

Binary

read_binary_files

Create a Dataset from binary files of arbitrary contents.

TFRecords

read_tfrecords	Create a Dataset from TFRecord files that contain tf.train.Example messages.
Dataset.write_tfrecords	Write the Dataset to TFRecord files.
TFXReadOptions	Specifies read options when reading TFRecord files with TFX.

Pandas

from_pandas	Create a Dataset from a list of pandas dataframes.
from_pandas_refs	Create a Dataset from a list of Ray object references to pandas dataframes.
Dataset.to_pandas	Convert this Dataset to a single pandas DataFrame.
Dataset.to_pandas_refs	Converts this Dataset into a distributed set of Pandas dataframes.

NumPy

read_numpy	Create an Arrow dataset from numpy files.
from_numpy	Creates a Dataset from a list of NumPy ndarrays.
from_numpy_refs	Creates a Dataset from a list of Ray object references to NumPy ndarrays.
Dataset.write_numpy	Writes a column of the Dataset to .npy files.
Dataset.to_numpy_refs	Converts this Dataset into a distributed set of NumPy ndarrays or dictionary of NumPy ndarrays.

Arrow

from_arrow	Create a Dataset from a list of PyArrow tables.
from_arrow_refs	Create a Dataset from a list of Ray object references to PyArrow tables.
Dataset.to_arrow_refs	Convert this Dataset into a distributed set of PyArrow tables.

MongoDB

read_mongo	Create a Dataset from a MongoDB database.
Dataset.write_mongo	Writes the Dataset to a MongoDB database.

BigQuery

read_bigquery(project_id[, dataset, query, ...])	Create a dataset from BigQuery.
Dataset.write_bigquery(project_id, dataset)	Write the dataset to a BigQuery dataset table.

SQL Databases

read_sql	Read from a database that provides a Python DB API2-compliant connector.
Dataset.write_sql	Write to a database that provides a Python DB API2-compliant connector.

Databricks

read_databricks_tables

Read a Databricks unity catalog table or Databricks SQL execution result.

Delta Sharing

read_delta_sharing_tables

Read data from a Delta Sharing table.

Hudi

read_hudi

Create a Dataset from an Apache Hudi table.

Iceberg

read_iceberg

Create a Dataset from an Iceberg table.

Lance

read_lance

Create a Dataset from a Lance Dataset.

Dask

from_dask	Create a Dataset from a Dask DataFrame.
Dataset.to_dask	Convert this Dataset into a Dask DataFrame.

Spark

from_spark	Create a Dataset from a Spark DataFrame.
Dataset.to_spark	Convert this Dataset into a Spark DataFrame.

Modin

from_modin	Create a Dataset from a Modin DataFrame.
Dataset.to_modin	Convert this Dataset into a Modin DataFrame.

Mars

from_mars	Create a Dataset from a Mars DataFrame.
Dataset.to_mars	Convert this Dataset into a Mars DataFrame.

Torch

from_torch

Create a Dataset from a Torch Dataset.

Hugging Face

from_huggingface

Create a MaterializedDataset from a Hugging Face Datasets Dataset or a Dataset from a Hugging Face Datasets IterableDataset.

TensorFlow

from_tf

Create a Dataset from a TensorFlow Dataset.

WebDataset

read_webdataset

Create a Dataset from WebDataset files.

Datasource API

read_datasource	Read a stream from a custom Datasource.
Datasource	Interface for defining a custom Dataset datasource.
ReadTask	A function used to read blocks from the Dataset.
datasource.FilenameProvider	Generates filenames when you write a Dataset.

Datasink API

Dataset.write_datasink	Writes the dataset to a custom Datasink.
Datasink	Interface for defining write-related logic.
datasource.RowBasedFileDatasink	A datasink that writes one row to each file.
datasource.BlockBasedFileDatasink	A datasink that writes multiple rows to each file.
datasource.FileBasedDatasource	File-based datasource for reading files.

Partitioning API

datasource.Partitioning	Partition scheme used to describe path-based partitions.
datasource.PartitionStyle	Supported dataset partition styles.
datasource.PathPartitionParser	Partition parser for path-based partition formats.
datasource.PathPartitionFilter	Partition filter for path-based partition formats.

MetadataProvider API

datasource.FileMetadataProvider	Abstract callable that provides metadata for the files of a single dataset block.
datasource.BaseFileMetadataProvider	Abstract callable that provides metadata for FileBasedDatasource implementations that reuse the base prepare_read() method.
datasource.DefaultFileMetadataProvider	Default metadata provider for FileBasedDatasource implementations that reuse the base prepare_read method.
datasource.ParquetMetadataProvider	Provides block metadata for Arrow Parquet file fragments.
datasource.FastFileMetadataProvider	Fast Metadata provider for FileBasedDatasource implementations.