VISSL

VISSL

  • Tutorials
  • Docs
  • GitHub

›

Tutorials

  • Overview

Getting Started

  • Installation
  • Understanding VISSL Training and YAML Config

Training

  • Train SimCLR on 1-gpu

Feature Extraction

  • Feature Extraction

Benchmark

  • Benchmark Linear Image Classification on ImageNet-1K
  • Benchmark Full-Finetuning on ImageNet-1K

Large Scale Training

  • Large Scale Training with VISSL (fp16, LARC, ZeRO, etc)

Inference

  • Using a pretrained model for inference

Welcome to VISSL Tutorials

These tutorials will help you understand how to use VISSL from examples which are in the form of ipython notebooks.

Run tutorials interactively

Each tutorial can be run interactively in Google Colaboratory which allows running the code directly in browser with access to GPUs. To run the tutorial in Colab, simply click on the button "Run in Google Colab" which looks like this:

Run in Google Colab

Every tutorial is standalone meaning that tutorial contain instructions for accessing data as well. At the start of every tutorial, the installation instructions are provided. We recommend to follow the tutorial steps to get started.

Run locally

There is also a button to download the notebook and source code to run it locally.

vissl
Facebook Open Source
Copyright © 2022 Facebook Inc
Legal:PrivacyTerms