sktime-tsfresh

What problem does it solve?

This Skill transforms raw time series data into tabular features, making it suitable for classical machine learning models, and automates feature extraction and selection to prevent data leakage.

Core Features & Use Cases

• Automated Feature Extraction: Leverages tsfresh to extract a wide array of statistical and spectral features from time series.
• Pipeline Integration: Seamlessly integrates tsfresh into sktime pipelines, ensuring feature extraction happens only on training data during cross-validation.
• Feature Relevance Selection: Uses the FRESH algorithm to filter out non-informative features, reducing dimensionality and improving model interpretability.
• Use Case: When building a time series classification model, use this skill to convert multiple sensor readings (time series) into a set of relevant features for a RandomForestClassifier.

Quick Start

Use the sktime-tsfresh skill to create a pipeline that extracts efficient tsfresh features and trains a RandomForestClassifier on the provided time series data X and labels y.