Skip to content

SparseML v1.4.2 Patch Release
Compare
Choose a tag to compare
@jeanniefinks jeanniefinks released this 09 Mar 21:28
· 10 commits to release/1.4 since this release
v1.4.2
1f50da6
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.

This is a patch release for 1.4.0 that contains the following changes:

  • Auto batch reduction for YOLOv5 on quantization aware training was failing with an OOM error. This has now been fixed and logic simplified to always scale the input batch size down by 4 and increase the gradient accumulation steps by 4 (neuralmagic/yolov5@2c7058d)
  • Older YOLOv5 sparse models, examples, and commands were failing to start in the 1.4 release due to the removal of hyps/hyp.finetune.yaml, data/hyps/hyp.finetune.yaml, models_v5.0/yolov5s.yaml files. These have now been added back in to ensure those training commands will still work and complete successfully. (neuralmagic/yolov5@5308d00, neuralmagic/yolov5@a1ebe6a, neuralmagic/yolov5@88aea22)
  • YOLOv5 models were crashing when trying to resume training from a checkpoint using the --resume arg. This is now fixed and --resume is supported again. (neuralmagic/yolov5@bd2bc0b)
  • Hugging Face token classification training CLI no longer fails when teacher and student labels are in different order. It now pulls the token configuration from the teacher model for the student. (#1400)
Assets 3
Loading