Ganesh Anand

Others are facing issues with the Docker as well. https://github.com/dmlc/decord/pull/166#issuecomment-1015189012

Hello, I have gotten this working @AntreasAntoniou @egorovivannn The issue was the path to which the libnvcuvid files are copied. `ln -s` also didn't work for me. I copied and...

I have been trying this too. @kcq This is my Dockerfile ```Dockerfile FROM nvcr.io/nvidia/pytorch:22.08-py3 WORKDIR / ADD run.py / CMD [ "python", "run.py" ] ``` and this is run.py ```python3...

Thank you. I will use it cautiously. I was able to get the above code working though with ```python import multiprocessing as mp mp.set_start_method('spawn', force=True) ``` at the beginning of...

Full script for converting without precision loss. I also used opset_version=17 as mentioned above. ```python import tensorrt as trt TRT_LOGGER = trt.Logger(trt.Logger.WARNING) def convert_onnx_to_trt(onnx_path, engine_path, batch_size=1, precision="fp16"): # Initialize TensorRT...

I'm seeing around 34ms on an RTX 2070 @mirza298 including Pillow opening of image. It could be 4ms faster with all fp16 layers and much faster if image loading is...

Try out a docker image like `nvcr.io/nvidia/tensorrt:24.09-py3`. @Wooho-Moon The repo mentions they used `10.4.0` which is in the above mentioned docker image. Might be some issue with your setup currently,...

I was also facing loss of precision. Have fixed it with this script based on ideas from other issues ```python import tensorrt as trt TRT_LOGGER = trt.Logger(trt.Logger.WARNING) def convert_onnx_to_trt(onnx_path, engine_path,...