Dheeraj Peri

1) We support aten::unbind operator now. So you could give it a try without fallback now. 2) If you still have some converter issues, can you try fallback with our...

test_dynamic_fallback has testcases where the model has dynamic shapes and ops that fallback. This issue is quite old and is related to dynamic shape aten::add not being converted properly. There...

@laikhtewari Here's the example for export : https://github.com/pytorch/TensorRT/blob/main/docsrc/user_guide/saving_models.rst

Hello @mfeliz-cruise @andi4191, have you seen this error or does it pass your test suite ? Internally, others also reported this converter test failure, so I wanted to double check.

Yes, this is a known issue and we brought it up with Pytorch. Investigation ongoing.

> `WARNING:torch_tensorrt.dynamo.conversion._ConverterRegistry:The converter for node aten.convolution.default received dynamic shaped inputs but the static version of the converter is being used. Please report this issue at https://github.com/pytorch/TensorRT/issues` > > This is...

> @peri044 I remember we've already removed `cudnn` dependency on the release/2.3, but it still stays here: > > https://github.com/pytorch/TensorRT/blob/3f6999d6ab2f9b62c63b78e8405cacf216370214/py/torch_tensorrt/__init__.py#L9 > > This causes an error when I import torch-trt:...

@ncomly-nvidia FX tests have been added in DLFW and this example works in the latest containers (verified in 22.11).

@tiny-dentist Has this bug been resolved with https://github.com/pytorch/TensorRT/issues/1363#issuecomment-1307618833 ?

@tiny-dentist The script you provided has `CarvanaDataset` which is missing. I've tried this testcase which is very similar to your usage or PTQ data loader calibrator https://github.com/pytorch/TensorRT/blob/main/tests/py/ptq/test_ptq_dataloader_calibrator.py. It passes on...