Ganesh Anand

VideoReader with cuda backend fails with num_workers > 1

2

### 🐛 Describe the bug I set the backend as CUDA. I have compiled torchvision master from source with ffmpeg 4.2.9 with nvenc. ```python import torchvision torchvision.set_video_backend("cuda") ``` in my...

Docker build fails

4

The Dockerfile in master doesn't work. ```bash /decord/src/video/ffmpeg/threaded_decoder.cc: In member function 'void decord::ffmpeg::FFMPEGThreadedDecoder::WorkerThread()': /decord/src/video/ffmpeg/threaded_decoder.cc:176:26: warning: catching polymorphic type 'struct dmlc::Error' by value [-Wcatch-value=] 176 | } catch (dmlc::Error error) {...

trtexec command gives batch_size 1 and loss of precision

4

Current trtexec command shown in the repo sets batch_size=1 even though onnx model is dynamic batch sized. I found this command helps export with dynamic batch size. ```bash trtexec --onnx=dfine_x_obj2coco.onnx...