Haihao Shen comments

Results 22 comments of


                                            Haihao Shen

Quantization AssertionError "inputs len must equal with input_tensor"

cc @chensuyue

How to De-age the face?

Is there any example to show how to de-age? It seems no related document.

Add INT8 Stable Diffusion through Optimum

@patrickvonplaten please review this one. Thanks.

Add INT8 Stable Diffusion through Optimum

> Hi @hshen14, > > Let's wait for `neural-compressor` and `optimum-intel` refactorization before increasing visibility ! Thanks @anton-l @echarlaix. Sure, let's do that.

Add INT8 Stable Diffusion through Optimum

Currently, Optimum-Intel was being upgraded with INC v2.0 API. Will re-visit this PR after the upgrade is done.

Add INT8 Stable Diffusion through Optimum

@echarlaix , do you think it's good time to revisit this? Thanks.

Add INT8 Stable Diffusion through Optimum

> Sure, I will work on it and open a PR on `diffusers` once everything is finalized, does that work for you @hshen14 ? That would work perfectly! Thanks @echarlaix

fixed the typos in the file

Thanks for the PR. We will merge it soon.

Fix typo "user_calibration_algos" in calibrator.py

Thanks for the PR. We will have a release with it by the end of this month.

can not conduct per-tensor quantisation?

> > Our design disables per-tensor weight selection to avoid exponential growth of the search space. You can check the file `neural_compressor/adaptor/pytorch_cpu.yaml` to add it back if needed. The best...