hexisyztem

I'll fix this detail in my next commit

This document is out of date > > There are no libcudadevrt.so and libcudart_static.so in your lib64 dir. > > I am confused with the cuda version, according to the...

> > document > > Thanks for your reply, i have build success and got a whl package as follows: > >  > > but, where i try to...

In addition, you can get the Bert model through lsi.Bert(...)

You can find interface code in this file: https://github.com/bytedance/lightseq/blob/master/lightseq/inference/pywrapper/wrapper.cc#L919

for example: import lightseq.inference as lsi model = lsi.Bert(model_file, max_batch_size) > In addition, you can get the Bert model through lsi.Bert(...)

It looks like you did compile correctly, it's just the wrong way to call it > > > > document > > > > > > > > > Thanks...

"Speedup of Transformer Inference" is not the speedup of vit model

here: https://github.com/bytedance/lightseq/blob/master/examples/inference/python/export/huggingface/hf_bart_export.py https://github.com/bytedance/lightseq/blob/master/examples/inference/python/test/ls_bart.py