For when you want an overview of a pandas DataFrame with significantly more depth than .describe(), try out https://github.com/pandas-profiling/pandas-profiling. Install with conda install -c conda-forge pandas-profiling or pip install pandas-profiling.
Some features of Pandas Profiling include:
- Essentials like types, unique values, missing values
- Quantile statistics
- Descriptive statistics
- Most frequent values
- Histogram
- Correlations highlighting of highly correlated variables
- Missing values matrix
For a useful library to help with explaining and interpreting models, try out EL5, short for "Explain it To Me Like I'm 5". Documentation is at https://eli5.readthedocs.io/en/latest and the source code is available from https://github.com/TeamHG-Memex/eli5. To install, do conda install -c conda-forge eli5 or pip install eli5.
- Add plotting functionality to scikit-learn objects with https://github.com/reiinakano/scikit-plot