Po-Wei (Vincent)
Po-Wei (Vincent)
@Shixiaowei02 any updates on this?
As said, TensorRT is mainly for inference optimizations. For training related issues please refer to each frameworks' approach. Thanks!
Might have mis-read. Thanks for the response. Let me check and get back to you.
Also, `pytorch_quantization` will not receive further development as stated [here](https://github.com/NVIDIA/TensorRT/tree/release/10.8/tools/pytorch-quantization). TensorRT-Model-Optimizer is now the encouraged path.
@symphonylyh any updates on this?
Seems like the `ReduceMin` is getting a scalar instead of a 1D tensor. Can you check if setting `keepdims=1` for the `ReduceSumSquare` OP works? See [ONNX spec for ReduceSumSquare](https://github.com/onnx/onnx/blob/main/docs/Operators.md#attributes-86) Another...
Closing this issue for now but feel free to reopen if you still see the problem!
> it does not work with my architecture There's many possible reasons that may lead to this. Could you provide your model for reference and maybe a brief description about...
Closing since no updates from requester after https://github.com/NVIDIA/TensorRT-LLM/pull/4030#issuecomment-2889886525. Feel free to reopen!
Closing since no response after https://github.com/NVIDIA/TensorRT-LLM/pull/2806#issuecomment-2746960894. Feel free to rebase and reopen if the PR is still relevant!