Welcome to DeepRank-GNN’s documentation!

DeepRank-GNN:

• Overview
  • 1) Graph generation
  • 2) Model Training
• Motivations
• Installation
  • Via Python Package
  • Via GitHub

Tutorial:

• Creating Graphs
  • Generate your graphs
  • Add your target values
  • Docking benchmark mode
• Training a module
  • 1. Select node and edge features
  • 2. Select the target (benchmarking mode)
  • 3. Select hyperparameters
  • 4. Load the network
  • 5. Train the model
  • 6. Analysis
  • 7. Save the model/network
  • 8. Test the model on an external dataset
• In short
• Use DeepRank-GNN paper’s pretrained model
• Advanced
  • Design your own GNN layer
  • Design your own neural network architecture
  • Use your GNN architecture in Deeprank-GNN

API

• Graph and Graph Generator
  • Graphs
  • Residue Graphs
  • Graph Generator
  • Parrallel Graph Generator
• Neural Network Training Evaluation and Test
  • Neural Network

Indices and tables

• Index

• Module Index

• Search Page