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

Pygcl

PyGcl is an open-source Python library built on PyTorch for implementing graph contrastive learning, with modular algorithm components, example methods, and utilities for training, evaluation and experiment management.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Pygcl — 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 PyGcl when building graph contrastive learning experiments in a Python/PyTorch environment and when reusable augmentation, objective, pairing and negative-mining components are useful.

Who it is for

  • Researchers prototyping graph representation learning methods.
  • Developers implementing or adapting graph contrastive learning algorithms in Python and PyTorch.

Documented uses

  • Construct augmented graph views for a contrastive-learning workflow.
  • Form positive and negative pairs from node or graph embeddings and compute contrastive objectives.
  • Apply negative-sample mining strategies.
  • Explore examples of published methods for node- and graph-level learning tasks.
  • Train, evaluate and manage graph-learning experiments.
Before you choose

Requirements & limits

The repository identifies PyGcl as open source under the Apache-2.0 license. The README documents installation through PyPI.

  • Requires a Python/PyTorch stack plus PyTorch-Geometric, DGL, scikit-learn, NumPy, tqdm and NetworkX.
  • DGL CPU and CUDA packages are separate; users must select and install the appropriate package themselves.
  • The library includes examples spanning several named methods, but the README does not specify results or dataset-specific performance.

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

Install PyGcl from PyPI

Install the library into a Python environment that has its stated dependencies.

Before you start

  • Python 3.8 or newer
  • PyTorch 1.9 or newer
  • PyTorch-Geometric 1.7
  • DGL 0.7 or newer
  • scikit-learn 0.24 or newer

Steps

  1. Install the listed prerequisites in the environment.

  2. Choose and install the appropriate DGL package for the intended CPU or CUDA setup.

  3. Run the command below.

    Documented command
    pip install PyGCL

  4. Import GCL from the environment.

Expected outcome

PyGcl is installed and its GCL module can be imported from the environment.

Guide limitations

  • The README does not provide a single DGL package choice for all environments; DGL must be installed separately because PyGcl does not check its version in setup.
  • Installation guidance does not establish that a particular environment or configuration has been tested.
Documentation 1 ↗
Evidence and transparency

Sources & review

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