Leo Dong comments

Results 22 comments of


                                            Leo Dong

agent.train --> TypeError: call() missing 1 required positional argument: 'network_state'

Same problem with td3... Any help is appreciated!

Drastically different results based on the number of query points

Just to clarify that the latent codes are exactly the same in both cases.

SDXL int8 Failure of TensorRT 10 when running txt2img_xl.py on GPU A100

Closed as duplicate of: https://github.com/NVIDIA/TensorRT-Model-Optimizer/issues/37

Concurrent inference failure of TensorRT 8.6.1 when running open_clip visual model tensorrt engine on GPU A100

@AmazDeng I think you didn't show how you hook the `TensorRTModel` to your Python multiprocessing pool. The issue is that a single TensorRT engine is not supposed to be held...

Concurrent inference failure of TensorRT 8.6.1 when running open_clip visual model tensorrt engine on GPU A100

> @LeoZDong > > Does TensorRT support multithreaded inference? Note that it's multithreading, not multiprocessing？ Yes. Look at the execute_async_v3 method (https://docs.nvidia.com/deeplearning/tensorrt/api/python_api/infer/Core/ExecutionContext.html#tensorrt.IExecutionContext.execute_async_v3). You will need to pass in a CUDA...

Tried to convert a mixed precision onnx model to TensorRT FP16 engine

Could you provide your ONNX model file?

No me permite instalar TENSORrt

Closing as stale.

add hunyuan dit txt2image

Seems like your converted ONNX graph has some issue in a reshape node. Could you attach the ONNX model file?

Error: Could not initialize EGL. and RuntimeError: Failed to initialize OpenGL

Same issue here...

convert to FP16 generate orphan and self-recurring nodes

@xiaowuhu Put up a fix here (#310). Mind assigning some reviewers when you get a chance? Thanks!