furkancoskun comments

Repositories
Issues
Comments

Results 3 comments of


                                            furkancoskun

Converting multi-input conv layer to tensorrt failed.

@weizhiyi777 @jinfagang @Xiaoyw1998 Have you solved the problem with Multi Input Convolution?

an inplace operation

I encountered the same problem. Is there anyone who solved it?

I solved the problem with adding `broadcast_buffers=False` to `torch.nn.parallel.DistributedDataParallel` [change main.py line 373 as following](https://github.com/antoyang/TubeDETR/blob/d70c0eed94da09a9a5e4329c937230ecfb1d3ee6/main.py#L373) ```python model = torch.nn.parallel.DistributedDataParallel( model, device_ids=[args.gpu], find_unused_parameters=True, broadcast_buffers=False, ) ```