SciKit-Learn Laboratory (SKLL)

SKLL (pronounced “skull”) provides a number of utilities to make it simpler to run common scikit-learn experiments with pre-generated features.

There are two primary means of using SKLL: the run_experiment script and the Python API.

Documentation

• Installation
• License
• Tutorial
  • Workflow
  • Titanic Example
    • Create virtual environment with SKLL
    • Get your data into the correct format
    • Create a configuration file for the experiment
    • Running your configuration file through run_experiment
    • Examine the results
  • IRIS Example on Binder
• Running Experiments
  • General Workflow
  • Feature files
    • arff
    • csv/tsv
    • jsonlines/ndj (Recommended)
    • libsvm
  • Configuration file fields
    • General
    • Input
    • Tuning
    • Output
  • Using run_experiment
  • Output files
    • Log files
    • Model files
    • Results files
    • Prediction files
    • Summary file
    • Folds file
    • Learning curve plots
  • Integration with Weights & Biases
• Using Custom Metrics
  • Writing Custom Metric Functions
  • Using in Configuration Files
  • Using via the API
• Utility Scripts
  • compute_eval_from_predictions
    • Positional Arguments
    • Optional Arguments
  • filter_features
    • Required Arguments
    • Optional Arguments
  • generate_predictions
    • Positional Arguments
    • Optional Arguments
  • join_features
    • Positional Arguments
    • Optional Arguments
  • plot_learning_curves
    • Positional Arguments
  • print_model_weights
    • Positional Arguments
    • Optional Arguments
  • skll_convert
    • Positional Arguments
    • Optional Arguments
  • summarize_results
    • Positional Arguments
    • Optional Arguments
• API Documentation
  • Quickstart
  • config Package
    • fix_json()
    • load_cv_folds()
    • locate_file()
  • data Package
    • data.featureset Module
    • data.readers Module
    • data.writers Module
  • experiments Package
  • learner Package
    • Learner Class
    • VotingLearner Class
  • metrics Module
    • correlation()
    • f1_score_least_frequent()
    • kappa()
    • register_custom_metric()
    • use_score_func()
  • utils Package
    • CLASSIFICATION_ONLY_METRICS
    • CORRELATION_METRICS
    • PROBABILISTIC_METRICS
    • REGRESSION_ONLY_METRICS
    • UNWEIGHTED_KAPPA_METRICS
    • WEIGHTED_KAPPA_METRICS
    • get_skll_logger()
  • types Module
    • ClassMap
    • ConfusionMatrix
    • FeatureDict
    • FeatureDictList
    • FeaturesetIterator
    • FoldMapping
    • IdType
    • IndexIterator
    • LabelType
    • LearningCurveSizes
    • FeatGenerator
    • PathOrStr
    • SparseFeatureMatrix
    • ComputeEvalMetricsResults
    • EvaluateTaskResults
    • CrossValidateTaskResults
    • VotingCrossValidateTaskResults
• Contributing
  • Guidelines
  • SKLL Code Overview
    • Organization
    • Entry Points & Workflow
• Internal Documentation
  • Release Process

Indices and tables

• Index

• Module Index

• Search Page