From 7220163968aab0ca0ff5f0b4700a9005ab589939 Mon Sep 17 00:00:00 2001 From: wenhuach21 Date: Fri, 7 Jun 2024 11:15:24 +0800 Subject: [PATCH] change cpu to auto_round format --- examples/language-modeling/main.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/examples/language-modeling/main.py b/examples/language-modeling/main.py index 8af4bbdc..588aa68a 100644 --- a/examples/language-modeling/main.py +++ b/examples/language-modeling/main.py @@ -329,7 +329,7 @@ def get_library_version(library_name): compression_dtype=torch.int8, compression_dim=0, use_optimum_format=False, device="xpu") if "cpu" in deployment_device: - autoround.save_quantized(output_dir=f'{export_dir}-cpu', format='itrex', inplace=inplace) + autoround.save_quantized(output_dir=f'{export_dir}-cpu', format='auto_round', inplace=inplace) if "fake" in deployment_device: model = model.to("cpu") model.save_pretrained(output_dir)