Merge branch 'main' into gen_embedding

.github/workflow_scripts/lint_check.sh

@@ -4,6 +4,7 @@ cd ../../
 set -ex
 
 python3 -m pip install --upgrade prospector pip
+yes | pip3 install astroid==v3.0.0
 FORCE_CUDA=1 python3 -m pip install -e '.[test]'  --no-build-isolation
 pylint --rcfile=./tests/lint/pylintrc ./python/graphstorm/data/*.py
 pylint --rcfile=./tests/lint/pylintrc ./python/graphstorm/dataloading/

docs/source/scale/sagemaker.rst

@@ -160,6 +160,11 @@ Because we are using a three-partition OGB-MAG graph, we need to set the ``--ins
 
 The trained model artifact will be stored in the S3 location provided through the ``--model-artifact-s3`` argument. You can use the following command to check the model artifacts after the training completes.
 
+If you want to resume a saved model checkpoint to do model fine-tuning you can pass
+the S3 address of the model checkpoint through the ``--model-checkpoint-to-load``
+argument. For example by passing ``--model-checkpoint-to-load s3://mag-model/epoch-2/``,
+GraphStorm will initialize the model parameters with the model checkpoint stored in ``s3://mag-model/epoch-2/``.
+
 .. code-block:: bash
 
     aws s3 ls s3://<PATH_TO_SAVE_TRAINED_MODEL>/

examples/mag/mag_gnn_nc.yaml

@@ -19,6 +19,8 @@ gsf:
     backend: gloo
     verbose: false
     save_perf_results_path: null
+  lm:
+    cache_lm_embed: true
   gnn:
     model_encoder_type: rgcn
     num_layers: 2