Closed xhangHU closed 2 years ago
Hi, could you please check where the code stucked by adding some debugging output?
After debugging, I found that it runs to 96 lines with torch.no_grad(): The code after that did not run
I think it’s stuck on 98 and 99 line
outputs, refined_depth, prob_maps = model(sample, cas_depth_num, cas_interv_scale, mode=args[[est_depth_1, pair_results_1], [est_depth_2, pair_results_2], [est_depth_3, pair_results]] = outputs
Finally I determined the location of the problem Is line 27 of model_cas.py out = self.init_conv(x) I am stuck at this step using my own data and the provided data set. How can I solve this problem?
I solved this problem. The reason is that multiple GPUs are used. When I use OMP_NUM_THREADS=1 MKL_NUM_THREADS=1 Solved the problem after restricting threads
When I ran the test, he was stuck at this step and there was no change for two days.
Number of model parameters: 1162696 load pretrained_model/vis/-1 0%| | 0/98 [00:00<?, ?it/s]