Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix sdxl q_unet config #2081

Merged
merged 3 commits into from
Dec 6, 2024
Merged

Fix sdxl q_unet config #2081

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
bb8751f
fix unet config
Kaihui-intel Dec 4, 2024
156f9b0
fix config attr setting
Kaihui-intel Dec 5, 2024
40a7ba6
fix accuracy script
Kaihui-intel Dec 5, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 1 addition & 7 deletions .../3.x_api/pytorch/diffusion_model/diffusers/stable_diffusion/smooth_quant/run_benchmark.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,19 +50,13 @@ function run_benchmark {

if [[ ${mode} == "performance" ]]; then
extra_cmd=$extra_cmd" --performance"
if [[ ${int8} == "true" ]]; then
extra_cmd=$extra_cmd" --int8"
fi
echo $extra_cmd

python -u sdxl_smooth_quant.py \
--model_name_or_path ${model_name_or_path} \
--latent ${latent} \
${extra_cmd}
else
if [[ ${int8} == "true" ]]; then
extra_cmd=$extra_cmd" --int8"
fi
echo $extra_cmd

python -u sdxl_smooth_quant.py \
Expand All @@ -82,7 +76,7 @@ function run_benchmark {
cd mlperf_sd_inference
cp ../main.py ./
if [ -d "../saved_results/" ]; then
mv ../saved_results/ ./
cp -r ../saved_results/ ./
fi

python -u main.py \
Expand Down
4 changes: 2 additions & 2 deletions ..._api/pytorch/diffusion_model/diffusers/stable_diffusion/smooth_quant/sdxl_smooth_quant.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -406,10 +406,10 @@ def forward_loop(model):
if args.int8:
from neural_compressor.torch.quantization import load
q_unet = load(os.path.abspath(os.path.expanduser(args.output_dir)))
setattr(q_unet, "config", pipeline.unet.config)
else:
q_unet = pipeline.unet

if not hasattr(q_unet, "config"):
setattr(q_unet, "config", pipeline.unet.config)
pipeline.unet = q_unet
quant_images = prompts2images(pipeline, prompts, n_steps=args.n_steps, latent=init_latent)
save_images(prompts, quant_images, args.output_dir, prefix='quant')
Expand Down
Loading