dataframe-api-compat 0.1.12

Implementation of the DataFrame Standard for pandas and Polars

DataFrame API Compat

Implementation of the DataFrame Standard for pandas and polars-eager

Note: there is ongoing discussion about lazy engines in the Standard. Until that has been resolved, this package should not be relied upon for polars-lazy.

What's this?

Please read our blog post! https://data-apis.org/blog/dataframe_standard_rfc/.

Documentation

Please check https://data-apis.org/dataframe-api/draft/API_specification/index.html for the methods supported by the Consortium Dataframe Standard.

How to try this out

Here's an example of how you can try this out:

import polars as pl

df = pl.DataFrame({'a': [1,2,3]})
df_std = df.__dataframe_consortium_standard__()

The object df_std is a Standard-compliant DataFrame. Check the API Specification for the full list of methods supported on it.

Compliance with the Standard

This is mostly compliant. Notable differences:

• for pandas numpy dtypes, the null values (NaN) don't follow Kleene logic;
• for polars lazy, column reductions (e.g. column.mean()) are not implemented;
• for polars lazy, comparisons between different dataframes are not implemented.

Installation

pip install dataframe-api-compat