Skip to content

0.5.0 - DDP , NLcodec + NLDb, scaling to large datasets

Compare
Choose a tag to compare
@thammegowda thammegowda released this 30 Mar 03:30
· 327 commits to master since this release
  • DDP: multinode training see scripts/slurm-multinode-launch.sh
  • FP16 and mixed-precision (upgrade from APEX to torch's built in AMP)
  • NLCodec & NLDb integration for scaling to large datasets using pyspark backend
  • Web UI rtg-serve
  • Cache ensemble state for rtg-decode
  • Docker images for 500-eng model
  • Parent-child transfer: Shrink parent model vocab and embeddings to child datasets
  • Fix packaging of flask app: now templates and static files are also included in PyPI package