Skip to content

Edge Feature Quantization [3/5]: Read and Partition - #751

Open
jchmura-sc wants to merge 9 commits into
mainfrom
jchmura/edge_feature_quantization_read_path
Open

Edge Feature Quantization [3/5]: Read and Partition#751
jchmura-sc wants to merge 9 commits into
mainfrom
jchmura/edge_feature_quantization_read_path

Conversation

@jchmura-sc

@jchmura-sc jchmura-sc commented Aug 15, 2026

Copy link
Copy Markdown
Collaborator

Summary

  • Reads packed edge features and their quantization metadata from TFRecords.
  • Carries quantized edge tensors through dataset construction and partitioning.
  • Keeps this layer independent of distributed loader materialization.

Stack

  1. Edge Feature Quantization [1/5]: Metadata Proto #749 - metadata proto
  2. Edge Feature Quantization [2/5]: Preprocessor #750 - preprocessor
  3. Read path and partitioning - this PR
  4. Distributed loading and materialization
  5. User guide

Comment thread tests/test_assets/distributed/run_distributed_partitioner.py Outdated
Comment thread gigl/common/data/load_torch_tensors.py Outdated
@jchmura-sc jchmura-sc self-assigned this Aug 17, 2026
Base automatically changed from jchmura/edge_feature_quantization_preprocessor to main August 18, 2026 06:01

@kmontemayor2-sc kmontemayor2-sc left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks Jacob!

Comment thread gigl/types/graph.py
Comment thread gigl/distributed/dataset_factory.py
Comment thread gigl/common/data/load_torch_tensors.py
Comment thread gigl/distributed/dist_range_partitioner.py Outdated
Comment thread gigl/distributed/dist_partitioner.py
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants