# skorch-dev/skorch

A scikit-learn compatible neural network library that wraps PyTorch

Repository: https://github.com/skorch-dev/skorch
Canonical: https://ross.abutalabs.com/products/skorch
Language: Jupyter Notebook
License: BSD-3-Clause
License Family: permissive
Topics: scikit-learn, pytorch, machine-learning, hacktoberfest, huggingface
Last push: 2026-08-10T09:21:34+00:00

## Health v2 (maintenance only)
Score: 85/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 97, release rhythm 60, longevity 100
- inputs: {"age_days": 3334, "days_push": 23, "days_rel": 111, "gap_med": 122.5, "n_releases_24m": 5}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 6173, forks 417 (observed 2026-08-28T04:09:36.991240+00:00)

## What it is
skorch is a Python library that wraps PyTorch neural networks in a scikit-learn compatible API, providing estimators like NeuralNetClassifier and NeuralNetRegressor. It lets you use sklearn pipelines, grid search, and cross-validation directly with PyTorch models.

## Use cases
- train pytorch models with scikit-learn api
- use pytorch neural network in sklearn pipeline
- grid search hyperparameters of a pytorch model
- cross-validate a neural network with sklearn
- wrap huggingface transformer in sklearn estimator
- replace keras-style fit loop with sklearn-compatible classifier

## When to choose
- you already use scikit-learn and want PyTorch models to fit into pipelines and grid search
- you want a simple fit/predict training loop without writing boilerplate PyTorch training code
- you need sklearn-compatible estimators for neural networks, including Hugging Face models

## When to avoid
- you need fine-grained control over every step of the PyTorch training loop
- you are building large-scale or highly customized deep learning training pipelines
- you do not use scikit-learn conventions and prefer raw PyTorch

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning
- domain: machine-learning, deep-learning, data-science
- platform: python
- tags: scikit-learn, pytorch, neural-networks, huggingface, grid-search

## Member repositories
- skorch-dev/skorch (main) score 85

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:36.991240+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:47:27.166695+00:00, confidence not recorded.
  - readme: https://github.com/skorch-dev/skorch (fetched 2026-08-28T04:09:36.991240+00:00, sha 0cc3334991e2)
- Data as of 2026-08-30T08:39:29.467469+00:00.