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 <quic_hitameht@quicinc.com>
Co-authored-by: Hitarth Mehta <quic_hitameht@quicinc.com>

updated release notes with known issue for Keras accuracy drop (#5454…

…) (#5455)

Signed-off-by: shobitha <shobitha@qti.qualcomm.com>

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 <quic_bhushans@quicinc.com>
Co-authored-by: Bhushan Sonawane <quic_bhushans@quicinc.com>

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 <quic_ykota@quicinc.com>
Co-authored-by: yathindra kota <quic_ykota@quicinc.com>

[r2.8 cherrypick] Use cudnn_conv_algo_search='EXHAUSTIVE' in ORT>1.19

Signed-off-by: Kyunggeun Lee <quic_kyunggeu@quicinc.com>
Co-authored-by: Kyunggeun Lee <quic_kyunggeu@quicinc.com>

Add Omniquant doc

Signed-off-by: Kevin Hsieh <klhsieh@qti.qualcomm.com>

Update release notes

Signed-off-by: Kevin Hsieh <klhsieh@qti.qualcomm.com>

images on pypi are not rendered correctly due to relative link to rep…

…o (#4890)

uploaded images instead to AWS and reference from there

resolves #4710

Signed-off-by: Bhushan Sonawane <quic_bhushans@quicinc.com>
Co-authored-by: Bhushan Sonawane <quic_bhushans@quicinc.com>

fix repo-sync and release workflow calls (#4804)

Signed-off-by: Bhushan Sonawane <quic_bhushans@quicinc.com>
Co-authored-by: Bhushan Sonawane <quic_bhushans@quicinc.com>

version 2.3.0

Release of the AI Model Efficiency toolkit package
Documentation main page: https://quic.github.io/aimet-pages/index.html