GiGL Components#

Config Populator

Component for processing template config files and updating with fields that are needed for downstream components.

Data Preprocessor

Component for reading and processing node, edge, and feature data and transforming it as needed for downstream components.

Subgraph Sampler

Component that generates k-hop localized subgraphs for each node in the graph.

Split Generator

Component to split the data into training, validation, and test sets.

Trainer

Component to run distributed training either locally or on the cloud.

Inferencer

Compoennt that runs inference to generate output embeddings and/or predictions