This release adds a few nifty features to `pandera`, special thanks to mastersplinter and ralbertazzi:
- We now have official [documentation](https://pandera.readthedocs.io/en/latest/)! Thanks to mastersplinter on the work here.
- the `Check` class now has a `groupby` argument, which enables the user to assert properties on subsets of the `Column` of interest. This opens up the possibility to compare the values or aggregates of values of subsets of a column 42.
- the introduction of hypothesis tests through the `Hypothesis` class, which is a subclass of the `Check` class. This enables the user to run hypothesis tests on their dataframe as part of a `DataFrameSchema` definition. Refer to the [documentation](https://pandera.readthedocs.io/en/latest/hypothesis.html) for more info #43.
- `Column`s now have a `required` argument (default = True), where `required=False` means that the column is optional 23.
- `SeriesSchemaBase` now has an `allow_duplicates` argument (default = True) 24
- add informative errors to `check_input` and `check_output` decorators 902f1990c3f0ebc07074faee3f3c7e123600a872
- `DataFrameSchema(..., strict=True)` means that all columns in the dataframe need to be specified in the schema `columns`. 34
- improved error messaging in general.
- improved CI (codecoverage).