johnzlli

pooling非对称pad支持

2

在下面这种结构中，pooling层需要非对称的pad，但是tnn转出来的模型只能支持对称pad。 

🐛 [Bug] Encountered bug when using Torch-TensorRT(torch.nn.TransformerEncoder)

2

## Bug Description Encountered error as follow when using Torch-TensorRT to convert torch.nn.LSTM in docker image nvcr.io/nvidia/pytorch:23.12-py3 :  To Reproduce ## To Reproduce example code: ```python import torch import...

🐛 [Bug] Encountered bug when using Torch-TensorRT(torch.nn.LSTM)

2

## Bug Description Encountered error as follow when using Torch-TensorRT to convert torch.nn.LSTM in docker image nvcr.io/nvidia/pytorch:23.12-py3 : NotImplementedError: aten::_cudnn_rnn_flatten_weight: attempted to run this operator with Meta tensors, but there...

❓ [Question] mlp running with torch_tensorrt slower than with inductor？

3

## ❓ Question I am within the nvcr.io/nvidia/pytorch:23.12-py3 container. The performance of torch_tensorrt is wrose than inductor. Details: example code ```python import torch import torch_tensorrt import torch.nn as nn class...

🐛 [Bug] Encountered bug for nn.LSTM with half dtype

2

## Bug Description When running the compiled LSTM model for half dtype with torch-tensorrt, I get this errors: `RuntimeError: Input and parameter tensors are not the same dtype, found input...