Sklearn Pipeline icon

Sklearn Pipeline

Create and use sklearn pipelines chaining transformers and estimators

Actions3

Train

AI-generated

Summary

Train a scikit-learn pipeline on the provided dataset.

Inputs

  • featureColumns (required) — Comma-separated list of feature column names to use.
  • targetColumn — Name of the target column; leave empty for transformer‑only pipelines.
  • preprocessors — Preprocessing steps to include in the pipeline.
  • estimator — Final estimator in the pipeline.
  • pcaComponents — Number of PCA components (if PCA is selected).
  • polyDegree — Polynomial degree (if Polynomial Features is selected).
  • pythonPath — Path to Python executable with scikit-learn installed.

Output shape

a JSON object containing the base64‑encoded serialized pipeline, the list of pipeline steps, feature column names, and optionally a training score

The pipeline is recreated on the Python side; successful execution returns this object, otherwise an error is raised.

Examples

Example 1: Train a classification pipeline on customer data

featureColumns: 'age,income'; targetColumn: 'churn'; preprocessors: ['StandardScaler']; estimator: 'LogisticRegression'

Example 2: Fit a regression pipeline with preprocessing

featureColumns: 'year,price'; estimator: 'RandomForestRegressor'; preprocessors: ['MinMaxScaler']; pcaComponents: 3

Example 3: Create a transformer‑only pipeline

featureColumns: 'sepal_length,sepal_width'; targetColumn: ''; preprocessors: ['StandardScaler']; estimator: 'none'

Discussion