After training I got this :
2021-07-02 16:20:58,583 maskrcnn_benchmark.inference INFO: Start evaluation on coco_2018_val dataset(209 images). 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 53/53 [00:13<00:00, 4.00it/s] 2021-07-02 16:21:11,846 maskrcnn_benchmark.inference INFO: Total run time: 0:00:13.263094 (0.06345977965724525 s / img per device, on 1 devices) 2021-07-02 16:21:11,846 maskrcnn_benchmark.inference INFO: Model inference time: 0:00:11.748324 (0.05621207844127308 s / img per device, on 1 devices) 2021-07-02 16:21:11,872 maskrcnn_benchmark.inference INFO: Preparing results for COCO format 2021-07-02 16:21:11,872 maskrcnn_benchmark.inference INFO: Preparing bbox results 2021-07-02 16:21:11,884 maskrcnn_benchmark.inference INFO: Preparing segm results 0it [00:00, ?it/s] Traceback (most recent call last): File "tools/train_net.py", line 222, in <module> main() File "tools/train_net.py", line 218, in main run_test(cfg, model, args.distributed) File "tools/train_net.py", line 139, in run_test output_folder=output_folder, File "/home/zhefei/maskrcnn-benchmark/maskrcnn_benchmark/engine/inference.py", line 115, in inference **extra_args) File "/home/zhefei/maskrcnn-benchmark/maskrcnn_benchmark/data/datasets/evaluation/__init__.py", line 22, in evaluate return coco_evaluation(**args) File "/home/zhefei/maskrcnn-benchmark/maskrcnn_benchmark/data/datasets/evaluation/coco/__init__.py", line 20, in coco_evaluation expected_results_sigma_tol=expected_results_sigma_tol, File "/home/zhefei/maskrcnn-benchmark/maskrcnn_benchmark/data/datasets/evaluation/coco/coco_eval.py", line 47, in do_coco_evaluation coco_results["segm"] = prepare_for_coco_segmentation(predictions, dataset) File "/home/zhefei/maskrcnn-benchmark/maskrcnn_benchmark/data/datasets/evaluation/coco/coco_eval.py", line 137, in prepare_for_coco_segmentation for mask in masks File "/home/zhefei/maskrcnn-benchmark/maskrcnn_benchmark/data/datasets/evaluation/coco/coco_eval.py", line 137, in <listcomp> for mask in masks IndexError: too many indices for tensor of dimension 3
❓ Questions and Help
After training I got this :
2021-07-02 16:20:58,583 maskrcnn_benchmark.inference INFO: Start evaluation on coco_2018_val dataset(209 images). 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 53/53 [00:13<00:00, 4.00it/s] 2021-07-02 16:21:11,846 maskrcnn_benchmark.inference INFO: Total run time: 0:00:13.263094 (0.06345977965724525 s / img per device, on 1 devices) 2021-07-02 16:21:11,846 maskrcnn_benchmark.inference INFO: Model inference time: 0:00:11.748324 (0.05621207844127308 s / img per device, on 1 devices) 2021-07-02 16:21:11,872 maskrcnn_benchmark.inference INFO: Preparing results for COCO format 2021-07-02 16:21:11,872 maskrcnn_benchmark.inference INFO: Preparing bbox results 2021-07-02 16:21:11,884 maskrcnn_benchmark.inference INFO: Preparing segm results 0it [00:00, ?it/s] Traceback (most recent call last): File "tools/train_net.py", line 222, in <module> main() File "tools/train_net.py", line 218, in main run_test(cfg, model, args.distributed) File "tools/train_net.py", line 139, in run_test output_folder=output_folder, File "/home/zhefei/maskrcnn-benchmark/maskrcnn_benchmark/engine/inference.py", line 115, in inference **extra_args) File "/home/zhefei/maskrcnn-benchmark/maskrcnn_benchmark/data/datasets/evaluation/__init__.py", line 22, in evaluate return coco_evaluation(**args) File "/home/zhefei/maskrcnn-benchmark/maskrcnn_benchmark/data/datasets/evaluation/coco/__init__.py", line 20, in coco_evaluation expected_results_sigma_tol=expected_results_sigma_tol, File "/home/zhefei/maskrcnn-benchmark/maskrcnn_benchmark/data/datasets/evaluation/coco/coco_eval.py", line 47, in do_coco_evaluation coco_results["segm"] = prepare_for_coco_segmentation(predictions, dataset) File "/home/zhefei/maskrcnn-benchmark/maskrcnn_benchmark/data/datasets/evaluation/coco/coco_eval.py", line 137, in prepare_for_coco_segmentation for mask in masks File "/home/zhefei/maskrcnn-benchmark/maskrcnn_benchmark/data/datasets/evaluation/coco/coco_eval.py", line 137, in <listcomp> for mask in masks IndexError: too many indices for tensor of dimension 3How can I solve this? Thanks in advance.