Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Tags: quic/aimet

Tags

2.17.0

Toggle 2.17.0's commit message
Fix failing fp16 QDQ export tests

Signed-off-by: Michael Tuttle <[email protected]>
Co-authored-by: Michael Tuttle <[email protected]>

2.16.0

Toggle 2.16.0's commit message
Update release notes for 2.16.0 release (#5788) (#5789)

Signed-off-by: Abhi Khobare <[email protected]>
(cherry picked from commit ebf1d92)

2.15.1

Toggle 2.15.1's commit message
cherry pick adascale changes from develop (#5739)

* Changes for experimental feature AdaScale in aimet-onnx

Signed-off-by: Abhi Khobare <[email protected]>

* Updated release notes and docs links

Signed-off-by: Abhi Khobare <[email protected]>

---------

Signed-off-by: Abhi Khobare <[email protected]>
Co-authored-by: Abhi Khobare <[email protected]>

2.15.0

Toggle 2.15.0's commit message
Add dummy action to make it easy to play with GH workflow (#5684)

Signed-off-by: Bhushan Sonawane <[email protected]>
Co-authored-by: Bhushan Sonawane <[email protected]>

2.14.0

Toggle 2.14.0's commit message
[cherry-pick] add e2e accuracy tests for aimet-tf (#5640)

* add e2e accuracy tests for aimet-tf

Signed-off-by: Bhushan Sonawane <[email protected]>

---------

Signed-off-by: Bhushan Sonawane <[email protected]>
Co-authored-by: Bhushan Sonawane <[email protected]>

2.13.0

Toggle 2.13.0's commit message
Pin deepspeed to stabilize CI pipeline (#5602)

Signed-off-by: Michael Tuttle <[email protected]>
Signed-off-by: Bhushan Sonawane <[email protected]>
Co-authored-by: Michael Tuttle <[email protected]>

2.12.0

Toggle 2.12.0's commit message
Adjust weight scale for int32 bias overflow in W16A16 quantization

Automatically adjust weight scale for int32 bias overflow in INT16 quantization

Signed-off-by: Hitarth Mehta <[email protected]>
Co-authored-by: Hitarth Mehta <[email protected]>

2.11.0

Toggle 2.11.0's commit message
updated release notes with known issue for Keras accuracy drop (#5454…

…) (#5455)

Signed-off-by: shobitha <[email protected]>

2.10.0

Toggle 2.10.0's commit message
Work-around onnx version converter issue for models with external (#5…

…371)

weights

If onnx version converter fails, upgrade and map weights separately

resolves #5367

Signed-off-by: Bhushan Sonawane <[email protected]>
Co-authored-by: Bhushan Sonawane <[email protected]>

2.9.0

Toggle 2.9.0's commit message
AdaScale: Maintain LR per model and apply it

* Remove hardcoding to the first sample during loss computation
* Fix tuples being passed to loss func

Signed-off-by: yathindra kota <[email protected]>
Co-authored-by: yathindra kota <[email protected]>