Back to Dask

Load and Save Data with Dask DataFrames

docs/source/dataframe-create.rst

2026.3.06.0 KB
Original Source

Load and Save Data with Dask DataFrames

.. meta:: :description: Learn how to create DataFrames and store them. Create a Dask DataFrame from various data storage formats like CSV, HDF, Apache Parquet, and others.

You can create a Dask DataFrame from various data storage formats like CSV, HDF, Apache Parquet, and others. For most formats, this data can live on various storage systems including local disk, network file systems (NFS), the Hadoop Distributed File System (HDFS), Google Cloud Storage, and Amazon S3 (excepting HDF, which is only available on POSIX like file systems).

See the :doc:DataFrame overview page <dataframe> for more on dask.dataframe scope, use, and limitations and :doc:DataFrame Best Practices <dataframe-best-practices> for more tips and solutions to common problems.

API

The following functions provide access to convert between Dask DataFrames, file formats, and other Dask or Python collections.

.. currentmodule:: dask.dataframe

File Formats:

.. autosummary:: read_csv read_parquet read_hdf read_orc read_json read_sql_table read_sql_query read_sql read_table read_fwf from_array to_csv to_parquet to_hdf to_sql

Dask Collections:

.. autosummary:: from_delayed from_dask_array from_map dask.bag.core.Bag.to_dataframe DataFrame.to_delayed to_records to_bag

Pandas:

.. autosummary:: from_pandas DataFrame.from_dict

Other File Formats:

  • Snowflake <https://github.com/coiled/dask-snowflake>_
  • Bigquery <https://github.com/coiled/dask-bigquery>_
  • Delta Lake <https://github.com/dask-contrib/dask-deltatable>_

Creating

Read from CSV


You can use :func:`read_csv` to read one or more CSV files into a Dask DataFrame.
It supports loading multiple files at once using globstrings:

.. code-block:: python

   >>> df = dd.read_csv('myfiles.*.csv')

You can break up a single large file with the ``blocksize`` parameter:

.. code-block:: python

   >>> df = dd.read_csv('largefile.csv', blocksize=25e6)  # 25MB chunks

Changing the ``blocksize`` parameter will change the number of partitions (see the explanation on
:ref:`partitions <dataframe-design-partitions>`). A good rule of thumb when working with
Dask DataFrames is to keep your partitions under 100MB in size.

Read from Parquet

Similarly, you can use :func:read_parquet for reading one or more Parquet files. You can read in a single Parquet file:

.. code-block:: python

df = dd.read_parquet("path/to/mydata.parquet")

Or a directory of local Parquet files:

.. code-block:: python

df = dd.read_parquet("path/to/my/parquet/")

For more details on working with Parquet files, including tips and best practices, see the documentation on :doc:dataframe-parquet.

Read from cloud storage


Dask can read data from a variety of data stores including cloud object stores.
You can do this by prepending a protocol like ``s3://`` to paths
used in common data access functions like ``dd.read_csv``:

.. code-block:: python

   >>> df = dd.read_csv('s3://bucket/path/to/data-*.csv')
   >>> df = dd.read_parquet('gcs://bucket/path/to/data-*.parq')

For remote systems like Amazon S3 or Google Cloud Storage, you may need to provide credentials.
These are usually stored in a configuration file, but in some cases you may want to pass
storage-specific options through to the storage backend.
You can do this with the ``storage_options`` parameter:

.. code-block:: python

   >>> df = dd.read_csv('s3://bucket-name/my-data-*.csv',
   ...                  storage_options={'anon': True})
   >>> df = dd.read_parquet('gs://dask-nyc-taxi/yellowtrip.parquet',
   ...                      storage_options={'token': 'anon'})

See the documentation on connecting to :ref:`Amazon S3 <connect-to-remote-data-s3>` or
:ref:`Google Cloud Storage <connect-to-remote-data-gc>`.

Mapping from a function

For cases that are not covered by the functions above, but can be captured by a simple map operation, :func:from_map is likely to be the most convenient means for DataFrame creation. For example, this API can be used to convert an arbitrary PyArrow Dataset object into a DataFrame collection by mapping fragments to DataFrame partitions:

.. code-block:: python

import pyarrow.dataset as ds dataset = ds.dataset("hive_data_path", format="orc", partitioning="hive") fragments = dataset.get_fragments() func = lambda frag: frag.to_table().to_pandas() df = dd.from_map(func, fragments)

Dask Delayed


:doc:`Dask delayed<delayed>` is particularly useful when simple ``map`` operations aren’t sufficient to capture the complexity of your data layout. It lets you construct Dask DataFrames out of arbitrary Python function calls, which can be
helpful to handle custom data formats or bake in particular logic around loading data.
See the :doc:`documentation on using dask.delayed with collections<delayed-collections>`.

Storing
-------

Writing files locally

You can save files locally, assuming each worker can access the same file system. The workers could be located on the same machine, or a network file system can be mounted and referenced at the same path location for every worker node. See the documentation on :ref:accessing data locally <connect-to-remote-data-local>.

Writing to remote locations


Dask can write to a variety of data stores including cloud object stores.
For example, you can write a ``dask.dataframe`` to an Azure storage blob as:

.. code-block:: python

   >>> d = {'col1': [1, 2, 3, 4], 'col2': [5, 6, 7, 8]}
   >>> df = dd.from_pandas(pd.DataFrame(data=d), npartitions=2)
   >>> dd.to_parquet(df=df,
   ...               path='abfs://CONTAINER/FILE.parquet'
   ...               storage_options={'account_name': 'ACCOUNT_NAME',
   ...                                'account_key': 'ACCOUNT_KEY'}

See the :doc:`how-to guide on connecting to remote data <how-to/connect-to-remote-data>`.