search

NVlabs / STEP

STEP: Spatio-Temporal Progressive Learning for Video Action Detection. CVPR'19 (Oral)
246 stars 48 forks source link

Run Demo 2th time and not working #17

Open huy99ls01 opened 3 years ago

huy99ls01 commented 3 years ago

I run demo 1st time in google colab and it's working, but when i try 2th time the demo is not working. This is what I got:

Warning: If you want to use fp16, please apex with cuda support (https://github.com/NVIDIA/apex) and update pytorch to 1.0 Warning: If you want to use fp16, please apex with cuda support (https://github.com/NVIDIA/apex) and update pytorch to 1.0 Warning: If you want to use fp16, please apex with cuda support (https://github.com/NVIDIA/apex) and update pytorch to 1.0 Loading pretrain model from /content/drive/MyDrive/STEP/pretrained/ava_step.pth

Building I3D model... Building I3D head for global branch... Building I3D head for global branch... Building I3D head for global branch... Building I3D head for context branch... Datalist len: 978

THCudaCheck FAIL file=/content/drive/My Drive/STEP/external/maskrcnn_benchmark/csrc/cuda/ROIAlign_cuda.cu line=321 error=209 : no kernel image is available for execution on the device Traceback (most recent call last): File "demo.py", line 229, in <module> main() File "demo.py", line 122, in main history, _ = inference(args, conv_feat, context_feat, nets, args.max_iter, tubes) File "/content/drive/My Drive/STEP/utils/utils.py", line 48, in inference pooled_feat = nets['roi_net'](conv_feat[:, T_start:T_start+T_length].contiguous(), flat_tubes) File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 727, in _call_impl result = self.forward(*input, **kwargs) File "/content/drive/My Drive/STEP/models/networks.py", line 45, in forward tubes.view(-1, 5).detach()) File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 727, in _call_impl result = self.forward(*input, **kwargs) File "/content/drive/My Drive/STEP/external/maskrcnn_benchmark/roi_layers/roi_align.py", line 92, in forward input, rois, self.output_size, self.spatial_scale, self.sampling_ratio File "/content/drive/My Drive/STEP/external/maskrcnn_benchmark/roi_layers/roi_align.py", line 53, in forward output = _C.roi_align_forward(input, roi, spatial_scale, output_size[0], output_size[1], sampling_ratio) RuntimeError: cuda runtime error (209) : no kernel image is available for execution on the device at /content/drive/My Drive/STEP/external/maskrcnn_benchmark/csrc/cuda/ROIAlign_cuda.cu:321

ghost commented 3 years ago

Hey man, can you please share you colab notebook? Thanks