• pandas == 0.22.0

• chi2 (2 stars)
• ANOVA (2 stars)
• T-test (2 stars)
• IV (3 stars)
• KS (3 stars)

• Collinearity
  • TBD
• Multicolinearity
  • Variance Inflation Factor (3 stars)

• PSI (3 stars)
• Dataframe comparison (unit tests-covered)

• Numeric
• Numeric-Categorical
• String-Categorical
• Time

• Tukey's method
• z-test

• Residual threshold method
• Local outlier factor
• HiCS

• Continous
  • mean
  • truncated mean
  • median
  • bin-nan
• Categorical
  • most frequent class
  • stringify

• onehot_split

• Purity (need unit tests)
• Accuracy (need unit tests)

• Equal pupulation binning (3 stars)
• Equal value binning (3 stars)
• Monotonic binning
• ChiMerge

• Dummy (2 stars)
• WOE (2 stars)
• Tree leaves encoding

• Spectral clustering
• Subspace clustering
• Multi-sourced clustering
• Multi-aspect clustering
• Multi-task clustering

• AE + K-means
• AE + Spectral clustering
• AE + Subspace clustering

• BiGAN
• infoGAN
• AAE

Exploring/Summarize the data distribution

different processing methods for differnet types of data

• continuous: Data that can take on any value in an interval
• discrete: Data that can only take on integer values

Data that can only take on a specific set of values

• Binary: special case of categorical data, can only take two values

Categorical data that has an explicit ordering

• mean
• truncated mean
• weighted mean
• median
• outliers

• variance: N-1
• standard deviation
• range: min/max values
• percentiles
• Interquartile Range(IQR): 75th percentile - 25th percentile

• Boxplot
• Frequency table
• histogram
• density plot: kernal density estimate

• Mode: the most commonly category/value
• Expected value: similar as weighted mean
• Bar charts:The frequency or proportion for each category plotted as bars
• Pie charts:The frequency or proportion for each category plotted as wedges in a pie =======

• Entity embeddings