diff --git a/scripts/convert.py b/scripts/convert.py index 9223d49..408e77d 100644 --- a/scripts/convert.py +++ b/scripts/convert.py @@ -142,7 +142,6 @@ def quantize(models_name_or_path, model_type): model_output=model, per_channel=True, reduce_range=True, # should be the same as per_channel - activation_type=QuantType.QUInt8, weight_type=weight_type, optimize_model=False, diff --git a/scripts/requirements.txt b/scripts/requirements.txt index a6e43bd..9ebdc80 100644 --- a/scripts/requirements.txt +++ b/scripts/requirements.txt @@ -1,5 +1,5 @@ torch transformers optimum[onnxruntime]>=1.7.1 -onnxruntime==1.10.0 +onnxruntime tqdm