Skip to content

MLOPTPSU/lazygcn

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

LazyGCN

Installation

pip install requirement.txt

python setup.py develop

Usage

import lazygcn

Loading dataset

set dataset directory

os.environ["GNN_DATASET_DIR"] = "<your dataset directory>"

dataset = data.Dataset('pubmed')

Supported datasets:

Running a model

There are several predefined training class, which prepare data and create pre-defined models and optimizer.

  • Models: GCN, MLP, RecGCN [GraphSAGE, GAT]*
  • Layers: GConv, Linear, RecGConv
  • Minibatch: Full, Random, Split
  • NodeSampler: LayerWise, LADIES, Exact, FullGraph [FastGCN, NodeWise]*

See notebooks folder for examples and lazygcn/training for pre-defined trainer.

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%