arnoegw

Hi Benedek, thanks again for sending this. This typo is no longer present on the main branch, so please let me close this pull request.

Catching up on old issues... It still is like @janpfeifer explained: broadcast and pool (the key operations to move data across GraphTensors) turn into `tf.gather` and `tf.unsorted_segment_`, and are executed...

Hi @SidneyLann, thanks for reaching out! What you describe is an inherent challenge of GNN modeling. I'll try and explain below, but with all due respect for the valid question:...

I believe this is no longer an issue - our Colab tutorials have been using `tfgnn.keras.*` for a long time now.

Sorry for the slow response. I believe this has been fixed soon after with commit https://github.com/tensorflow/gnn/commit/e6ce2ac37b6f2ce27e12fb61405d5797b7a7b500

I suppose this issue has been resolved, so I'm going to close it.

Commit https://github.com/tensorflow/gnn/commit/5fea463ea79c3ca7314517c60c605d4377cd4b7c has now made the fix at the source (keras/layers/graph_ops.py). The change to api_docs/python/tfgnn/keras/layers.md will happen at the next refresh of api_docs from the source. With that, this pull...