diff --git a/docs/source/api/notebooks/index.rst b/docs/source/api/notebooks/index.rst index 775a8ae9e..4d12b9a84 100644 --- a/docs/source/api/notebooks/index.rst +++ b/docs/source/api/notebooks/index.rst @@ -19,7 +19,7 @@ Once they're familiar with customizing components, users can refer to `Notebook These examples utilize GraphStorm APIs, such as ``graphstorm``, ``graphstorm.dataloading.GSgnnDataset``, ``graphstorm.dataloading.GSgnnNodeDataLoader``, ``graphstorm.trainer.GSgnnNodePredictionTrainer``, and ``graphstorm.inference.GSgnnNodePredictionInferrer``, to form the training and infernece pipeline. In terms of the GNN models, users can refer to the `demo_model.py `_ file in which all models are created by using GraphStorm APIs. -These notebooks can run with the GraphStrom Standalone mode, i.e., on a single CPU or GPU of a single machine. To fully leverage GraphStorm's distributed model training and inference capability, users can follow the guidelines shown in `Notebook 6: Converting Customized Model Notebooks to Using GraphStorm CLIs `_ and choose a proper distributed environment to run the custom models for their enterprise-level graph datasets. Users can refer to the `demo_run_train.py `_ and `demo_run_infer.py `_ as examples of the converted model Python files to be used by GraphStorm CLIs. +These notebooks can run with the GraphStrom Standalone mode, i.e., on a single CPU or GPU of a single machine. To fully leverage GraphStorm's distributed model training and inference capability, users can follow the guidelines shown in `Notebook 6: Running Custom Model with GraphStorm CLIs `_ and choose a proper distributed environment to run the custom models for their enterprise-level graph datasets. Users can refer to the `demo_run_train.py `_ and `demo_run_infer.py `_ as examples of custom models to be used by GraphStorm CLIs. .. toctree:: :maxdepth: 2