OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Torchssl

TorchSSL is an open-source Python toolbox built on PyTorch for implementing and comparing semi-supervised learning methods, including the official FlexMatch implementation.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Torchssl — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use TorchSSL when you need to inspect or work from its existing research implementations and experiment configurations, while accounting for its deprecated, unmaintained status.

Who it is for

  • Machine-learning researchers comparing semi-supervised learning algorithms.
  • Developers adapting an existing semi-supervised learning implementation for experiments.

Documented uses

  • Compare semi-supervised methods against a fully supervised baseline.
  • Configure experiments across supported image datasets and labeled-sample settings.
  • Add a custom algorithm using the repository's model, training-file and configuration-file structure.
Before you choose

Requirements & limits

The public repository specifies an MIT license.

  • The repository includes configuration files and reported results for semi-supervised learning experiments.
  • Its maintainers describe USB as a successor built upon TorchSSL, with a broader codebase and benchmark.
  • The repository warns that later code changes and merged pull requests will not be rerun to update results.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare the TorchSSL Conda environment

Clone the repository and initialize its documented Conda environment.

Before you start

  • Anaconda or Miniconda

Steps

  1. Clone the TorchSSL repository to your machine.

  2. Run conda env create -f environment.yml to initialize the environment.

Expected outcome

The repository's Conda environment is initialized for working with TorchSSL.

Guide limitations

  • The repository is deprecated and no longer maintained.
Documentation 1 ↗
Source-based practical guide

Configure and launch a FlexMatch experiment

Run the repository's documented FlexMatch experiment after initializing its environment.

Before you start

  • TorchSSL cloned locally
  • Anaconda or Miniconda and the initialized Conda environment

Steps

  1. Edit config/flexmatch/flexmatch.yaml to set the experiment configuration.

  2. Run the command below.

    Documented command
    python flexmatch.py --c config/flexmatch/flexmatch.yaml

Expected outcome

The documented FlexMatch experiment is launched with the selected configuration.

Guide limitations

  • TorchSSL is deprecated and no longer maintained; later code changes will not trigger updated results.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
16 October 2026
Original website title
TorchSSL/TorchSSL · Official homepage title · retained in its source language
How we review sourcesSuggest a correction

Related options

USB (Semi-supervised-learning) — The TorchSSL maintainers identify it as an upgrade and successor codebase built upon TorchSSL, with a broader benchmark spanning computer vision, natural language processing and au