Feather development lives on in Apache Arrow. The arrow R package includes a much faster implementation of Feather, i.e. arrow::read_feather. The Python package feather is now a wrapper around pyarrow.feather.

Feather provides binary columnar serialization for data frames. It is designed to make reading and writing data frames efficient, and to make sharing data across data analysis languages easy. This initial version comes with bindings for python (written by Wes McKinney) and R (written by Hadley Wickham).

Feather uses the Apache Arrow columnar memory specification to represent binary data on disk. This makes read and write operations very fast. This is particularly important for encoding null/NA values and variable-length types like UTF8 strings.

Feather is a part of the broader Apache Arrow project. Feather defines its own simplified schemas and metadata for on-disk representation.

Feather currently supports the following column types:

• A wide range of numeric types (int8, int16, int32, int64, uint8, uint16, uint32, uint64, float, double).
• Logical/boolean values.
• Dates, times, and timestamps.
• Factors/categorical variables that have fixed set of possible values.
• UTF-8 encoded strings.
• Arbitrary binary data.

All column types support NA/null values.

pip install feather-format

install.packages("feather")

julia> using Pkg
julia> Pkg.add("Feather")

This library is released under the Apache License, Version 2.0.

See NOTICE for details about the library's copyright holders.

To get started with the python bindings, see the python feather documentation

To get started with the R bindings, see the R feather documentation

To get started with the Julia bindings see Feather.jl