Skip to main content
Ctrl+K
skrub - Home skrub - Home
  • Install
  • User Guide
  • How-tos
  • API
  • Examples
    • Learning Materials
    • Release history
    • Development
  • GitHub
  • Discord
  • Bluesky
  • X (ex-Twitter)
  • Install
  • User Guide
  • How-tos
  • API
  • Examples
  • Learning Materials
  • Release history
  • Development
  • GitHub
  • Discord
  • Bluesky
  • X (ex-Twitter)

Section Navigation

  • Hands-On with Column Selection and Transformers
  • SquashingScaler: Robust numerical preprocessing for neural networks
  • Sessions in time-based data: Predicting user purchases with the SessionEncoder
  • Encoding features
    • Encoding: from a dataframe to a numerical matrix for machine learning
    • Various string encoders: a sentiment analysis example
    • Handling datetime features with the DatetimeEncoder
  • Skrub DataOps
    • Multiples tables: building machine learning pipelines with DataOps
    • Hyperparameter tuning with DataOps
    • Tuning DataOps with Optuna
    • Subsampling for faster development
    • Use case: developing locally and deploying to production
    • Using PyTorch (via skorch) in DataOps
  • Joining tables with imperfect data
    • Fuzzy joining dirty tables with the Joiner
    • Deduplicating misspelled categories
    • Spatial join for flight data: Joining across multiple columns
    • AggJoiner on a credit fraud dataset
    • Interpolation join: infer missing rows when joining two tables
  • Examples
  • Encoding features

Encoding features#

Encoding: from a dataframe to a numerical matrix for machine learning

Encoding: from a dataframe to a numerical matrix for machine learning

Various string encoders: a sentiment analysis example

Various string encoders: a sentiment analysis example

Handling datetime features with the DatetimeEncoder

Handling datetime features with the DatetimeEncoder

previous

Sessions in time-based data: Predicting user purchases with the SessionEncoder

next

Encoding: from a dataframe to a numerical matrix for machine learning

Show Source
Financial support from inria and :probabl.
Part of scikit-learn central