Catalyst logo

PyTorch framework for Deep Learning research and development.

It focuses on reproducibility, rapid experimentation, and codebase reuse so you can create something new rather than write another regular train loop. Break the cycle - use the Catalyst!

Project manifest. Part of PyTorch Ecosystem. Part of Catalyst Ecosystem:
  • Alchemy - experiments logging & visualization

  • Catalyst - accelerated deep learning R&D

  • Reaction - convenient deep learning models serving

Catalyst at AI Landscape.

Getting started

import os
import torch
from torch.nn import functional as F
from import DataLoader
from torchvision.datasets import MNIST
from torchvision.transforms import ToTensor
from catalyst import dl, metrics

model = torch.nn.Linear(28 * 28, 10)
optimizer = torch.optim.Adam(model.parameters(), lr=0.02)

loaders = {
    "train": DataLoader(MNIST(os.getcwd(), train=True, download=True, transform=ToTensor()), batch_size=32),
    "valid": DataLoader(MNIST(os.getcwd(), train=False, download=True, transform=ToTensor()), batch_size=32),

class CustomRunner(dl.Runner):

    def predict_batch(self, batch):
        # model inference step
        return self.model(batch[0].to(self.device).view(batch[0].size(0), -1))

    def _handle_batch(self, batch):
        # model train/valid step
        x, y = batch
        y_hat = self.model(x.view(x.size(0), -1))

        loss = F.cross_entropy(y_hat, y)
        accuracy01, accuracy03 = metrics.accuracy(y_hat, y, topk=(1, 3))
            {"loss": loss, "accuracy01": accuracy01, "accuracy03": accuracy03}

        if self.is_train_loader:

runner = CustomRunner()
# model training
# model inference
for prediction in runner.predict_loader(loader=loaders["valid"]):
    assert prediction.detach().cpu().numpy().shape[-1] == 10
# model tracing
traced_model = runner.trace(loader=loaders["valid"])

Step by step guide

  1. Start with Catalyst 101 — Accelerated PyTorch introduction.

  2. Go through Kittylyst if you would like to dive into the core design concepts of the framework.

  3. Check minimal examples.

  4. Try notebook tutorials with Google Colab.

  5. Read blogposts with use-cases and guides.

  6. Learn machine learning with our “Deep Learning with Catalyst” course.

  7. Or go directly to advanced classification, detection and segmentation pipelines.

  8. Want more? See Alchemy and Reaction packages.

  9. RL fan? Please follow Catalyst.RL repo.

  10. If you would like to contribute to the project, follow our contribution guidelines.

  11. If you want to support the project, feel free to donate on patreon page or write us with your proposals.

  12. Finally, do not forget to join our slack for collaboration.


Catalyst helps you write compact but full-featured Deep Learning pipelines in a few lines of code. You get a training loop with metrics, early-stopping, model checkpointing and other features without the boilerplate.


Common installation:

pip install -U catalyst

More specific with additional requirements:

pip install catalyst[cv]         # installs CV-based catalyst
pip install catalyst[nlp]        # installs NLP-based catalyst
pip install catalyst[ecosystem]  # installs Catalyst.Ecosystem
# and master version installation
pip install git+ --upgrade

Catalyst is compatible with: Python 3.6+. PyTorch 1.1+.

Tested on Ubuntu 16.04/18.04/20.04, macOS 10.15, Windows 10 and Windows Subsystem for Linux.


  • callbacks - a variety of callbacks for your train-loop customization.

  • contrib - additional modules contributed by Catalyst users.

  • core - framework core with main abstractions - Experiment, Runner and Callback.

  • data - useful tools and scripts for data processing.

  • dl - entrypoint for your deep learning experiments.

  • experiments - a number of useful experiments extensions for Notebook and Config API.

  • metrics – classic ML and CV/NLP/RecSys metrics.

  • registry - Catalyst global registry for Config API.

  • runners - runners extensions for different deep learning tasks.

  • tools - extra tools for Deep Learning research, class-based helpers.

  • utils - typical utils for Deep Learning research, function-based helpers.


All Catalyst code, features and pipelines are fully tested with our own catalyst-codestyle.

In fact, we train a number of different models for various of tasks - image classification, image segmentation, text classification, GANs training and much more. During the tests, we compare their convergence metrics in order to verify the correctness of the training procedure and its reproducibility.

As a result, Catalyst provides fully tested and reproducible best practices for your deep learning research.

Indices and tables