Matt Wittbrodt

fn.readers.video_resize at longer sequence lengths gives dali::CUDAError without nvidia-smi

12

Hello - I've been running into an issue where long sequence lengths in video_resize are crashing the readers.video_resize pipe. I am using a V100 (16160 MB memory) with the following...

ReadMe example missing a comma

1

model = ConvLSTM(input_dim=channels, hidden_dim=[64, 64, 128], kernel_size=(3, 3), num_layers=3, batch_first=True[missing comma here] bias=True, return_all_layers=False)

Version: 1.0.183 Platform: Windows x64 10.0.17763 Plugins: none Frontend: xterm-webgl Whenever using "interactive" applications (requiring screen refresh, etc), I get a blank screen:  After typing Ctrl+C to exit, part...