wehub/nvidia--tensorrt
1
0
Fork 0
Code Issues Pull Requests Actions Deployments Models Agent Notes Packages Projects Releases Wiki Activity

chore: import upstream snapshot with attribution
Docker Image CI / build-ubuntu2004 (push) Has been cancelled
Details

Browse Source
This commit is contained in:
wehub-resource-sync
2026-07-13 13:36:55 +08:00
commit c8a779b1bb
1887 changed files with 3245738 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/tensorflow-quantization/CHANGELOG.md
+16
View File
@@ -0,0 +1,16 @@
# NVIDIA QAT toolkit for Tensorflow change log
Dates are in YYYY-MM-DD format.
## v0.2.0 (2022-09-09)
- Fixed bug in `infer_engine.py` (low accuracy due to indentation error).
- Added Inception-v3 support: code and results.
- Fixed User Guide links (they were the internal links, not public ones).
- Fixed bug in `examples/mobilenet/test_qdq_node_placement.py` (we implemented a more general `get_tfkeras_model` for Keras models QAT workflows, but missed that change in the MobileNet test script).
- Added Pillow requirement in examples.
- Added note for Conv2DTranspose support in the main README file.
- Improved generalization of residual branch detection for QDQ node placement.
## v0.1.0 (2022-06-17)
- Initial release of the toolkit.