Xiaodong (Vincent) Huang comments

Results 283 comments of


                                            Xiaodong (Vincent) Huang

onnx convert to TensorRT failed

closing since no activity for more than 14 days, please reopen if you still have question, thanks!

TensorRT8 int8 model convtranspose layer output error

closing since no activity for more than 14 days, please reopen if you still have question, thanks!

3D asymmetric padding is not supported on pre-sm70 GPUs

closing since no activity for more than 14 days, please reopen if you still have question, thanks!

Get wrong quantization result after setting TensorQuantizer.use_fb_fake_quant = True

Thanks @Wang-Qk , the pytorch-quantization is tested with torch1.9.1+, see README in https://github.com/NVIDIA/TensorRT/tree/main/tools/pytorch-quantization, there might other issues for lower version torch.

Get wrong quantization result after setting TensorQuantizer.use_fb_fake_quant = True

Closing since no activity for more than 3 weeks, thank you!

Do I have to do PTQ before QAT with pytorch_quantization toolkit?

closing old issues that is inactive for long time, thanks all!

what does the errors mean?

closing since no activity for more than 3 weeks, please reopen if you still have question, thanks!

Is there a way to speed up collect_stats() method in pythorch quantization?

We will enable the cuda extension by default in next release, closing this issue, thanks!

trtexec uses half the memory of python API

closing since no activity for more than 3 weeks, please reopen if you still have question, thanks!

How to install Tensorrt8 on tx2

closing since no activity for more than 14 days, please reopen if you still have question, thanks!