Seth Michael Larson

Add support for simple slicing of DataFrames

1

Add support for: - `df[:]` - `df[::]` - `df[:100]` - `df[100:]` - `df[100:200]` - `df[100:][1:]` Slicing functionality that should raise a ValueError: - `df[::2]` (step) - Probably others?

May be a no-op, but we need to run our test suite with pandas 1.4.x to make sure.

Add a way to access _score from DataFrame when using scoring filters

1

Relates to #282 it'd be nice to be able to access the `_score` value (and sort by it too). Need to find out how we should expose the `_score` information...

Create a test case for using Eland with a datastream

2

Currently we don't test (or document) that Eland should work with data streams, we should probably test that everything works properly.

Pulled from: - https://github.com/elastic/eland/issues/118#issue-556293360 - https://github.com/elastic/eland/issues/118#issuecomment-590240334

Add support for pd.Grouper() in DataFrame.groupby()

1

- Alias `eland.Grouper()` to be the same as `pd.Grouper()` for convenience - Eventually will be used for `Grouper(..., freq="...")` with timestamp bucketing.

Implement DatetimeIndex

1

See [`pandas.DatetimeIndex`](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DatetimeIndex.html) and [`date_range`](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.date_range.html#pandas.date_range). End goal is to be able to visualize time-series data on a histogram.

Allow filtering on a GroupByDataFrame

1

Pretty sure pandas allows you to filter once you've called groupby(), we should as well.

Add TQDM progress bar for Jupyter Notebook pandas_to_eland() calls

2

- https://github.com/tqdm/tqdm#ipythonjupyter-integration - Units are rows/s - Maybe hide progress bar on completion? Need to test it out first. - Add an option `show_progressbar` which defaults to `None` - If...

Implement [DataFrame Series].fillna() and [DataFrame, Series].dropna()

1

`fillna()` can be implemented as a post-processing task. `dropna()` can be implemented as a query filter For now we can support `axis` only as `axis=1` for "columns", some options may...