Doc error on the comparison page - Dask does support a quantile method (albeit approximate)

[ianozsvald]

On https://modin.readthedocs.io/en/latest/comparisons/dask.html you note Dask is currently missing multiple APIs from pandas that Modin has implemented. Of note: Dask does not implement iloc, MultiIndex, apply(axis=0), quantile, median, and more.

Dask has a quantile method for both Series and DataFrame: https://docs.dask.org/en/latest/generated/dask.dataframe.DataFrame.quantile.html . I'd suggest either removing "quantile" from that sentence or maybe noting that only an approximate quantile is available.

I agree that an exact median is not supported, nor is that likely.

The approximate quantile method has been improved (details for the morbid: dask/dask#3819 (comment) ), the current default has better error characteristics than from several years back and a newer tdigest options allows for error guarantees.

[YarShev]

Hi @ianozsvald , thanks for the notice! We'll definitely correct this.