Classification

Content

Models

  • Logistic regression
  • Multi-layer perceptron
  • Decision tree
  • Random forest
  • Gradient boost tree
  • SVM
  • Naive Bayes

Evaluation Metrics

  • Accuracy
  • Precision
  • Recall
  • F1
  • AUROC
  • AUPR

Feature Normalization and Scaling

  • Standard scaling (z-score)
  • Min-max 0-1 scaling
  • Min-max -1-1 scaling
  • Max absolute scaling
  • L1 normalization
  • L2 normalization

Other Functionality

  • ROC and other curves for binary classification
  • Subsampling
  • Cross-validation (param grid)
  • Feature selection by p-value (ANOVA, Chi-square)
  • Dimensionality reduction by PCA
  • Results storable to DB
  • Queryable and exportable to a (new) data set