siddharth-sharma7 / fast-Bart

Convert BART models to ONNX with quantization. 3X reduction in size, and upto 3X boost in inference speed
34 stars 3 forks source link

TypeError: quantize_dynamic() got an unexpected keyword argument 'activation_type' #5

Open arnabmanna619 opened 2 years ago

arnabmanna619 commented 2 years ago

I am trying to replicate the code but I get an error related to quantization. I'm running it on Colab.

image

Gooogr commented 1 year ago

Faced with the same problem. Fixed issue by editing the requirements.txt file:

onnxruntime==1.7.0
dangcaptkd2 commented 1 year ago

onnxruntime==1.7.0 Thanks, this help me.

DYF-AI commented 11 months ago

You can comment the key 'activation_type' to solve this problem.