skorch-dev/skorch

A scikit-learn compatible neural network library that wraps PyTorch

What it solves

It bridges the gap between PyTorch and scikit-learn, allowing users to use PyTorch neural networks as if they were scikit-learn estimators. This eliminates the need to write repetitive boilerplate code for training loops, validation splits, and hyperparameter tuning when using PyTorch.

How it works

Skorch wraps PyTorch modules into a scikit-learn compatible interface. By providing a NeuralNetClassifier or similar wrapper, it enables the use of standard scikit-learn tools like Pipeline for preprocessing and GridSearchCV for automated hyperparameter optimization, while the underlying model remains a PyTorch neural network.

Who it’s for

Machine learning engineers and data scientists who want the flexibility of PyTorch for building models but prefer the streamlined, standardized workflow and utility tools provided by scikit-learn.

Highlights

  • Scikit-learn Compatibility: Full integration with Pipeline and GridSearchCV.
  • Training Utilities: Built-in support for early stopping, checkpointing, and learning rate schedulers.
  • Training Management: Includes progress bars for CLI and Jupyter and parameter freezing/unfreezing.
  • Ecosystem Integration: Compatible with Hugging Face and GPyTorch for Gaussian Processes.

Related

  • Project
    TorchSpatiotemporal/tsl

    A PyTorch-based library for neural spatiotemporal forecasting that provides a unified framework for data preprocessing, model prototyping, and Graph Neural Network implementation.

  • Project
    pykt-team/pykt-toolkit

    A PyTorch-based library for training and benchmarking deep learning knowledge tracing models with standardized data preprocessing and multiple model implementations.

  • Project
    BindsNET/bindsnet

    A PyTorch-based library for simulating spiking neural networks (SNNs) to develop biologically inspired machine learning and reinforcement learning algorithms.

  • Project
    skforecast/skforecast

    A Python library for time series forecasting that integrates scikit-learn compatible models, statistical methods, and foundation models into a unified forecasting workflow.

  • Project
    speechbrain/speechbrain

    An open-source PyTorch toolkit for Conversational AI that provides recipes and tools for speech and text processing, including speech recognition and synthesis.