TensorRT
TensorRT copied to clipboard
PyTorch/TorchScript/FX compiler for NVIDIA GPUs using TensorRT
Implement a converter for aten.as_strided https://pytorch.org/docs/stable/generated/torch.as_strided.html#torch-as-strided
Implement a converter for aten._cdist_forward https://pytorch.org/docs/stable/generated/torch.cdist.html#torch.cdist https://github.com/pytorch/pytorch/blob/f00ece024bd70e6a5f3aa763715a1f353cfdacdf/aten/src/ATen/native/Distance.cpp#L81
Can TensorRT load and run torch_tensorrt models directly? I want to export my pytorch model and deploy it with TensorRT.
This page has several issues with the provided examples: https://pytorch.org/TensorRT/user_guide/saving_models.html 1. In the examples, an optional parameter is passed before a non-optional parameter which is invalid python. This issue also...
## ❓ Question I tried compiling a few models with `torch_tensorrt.compile(model, inputs, ir='dynamo', ...)` and each one of them was slower than the respective uncompiled model. I was wondering if...
## - **Function Schema**: - **Original PyTorch API**: - **Relevant TensorRT Documentation**: ## Alternatives ## Additional context
# Description Adds converter support for aten::flip https://pytorch.org/docs/stable/generated/torch.flip.html Fixes # (issue) ## Type of change Please delete options that are not relevant and/or add your own. - Bug fix (non-breaking...