the information is as follow:
"
INFO:maskrcnn_benchmark.inference:Prepare ground truth annotations.
Traceback (most recent call last):
File "/home/jinyuda/.pycharm_helpers/pydev/pydevd.py", line 1483, in _exec
pydev_imports.execfile(file, globals, locals) # execute the script
File "/home/jinyuda/.pycharm_helpers/pydev/_pydev_imps/_pydev_execfile.py", line 18, in execfile
exec(compile(contents+"\n", file, 'exec'), glob, loc)
File "/home/jinyuda/scene_graph_benchmark/tools/train_sg_net.py", line 229, in
main()
File "/home/jinyuda/scene_graph_benchmark/tools/train_sg_net.py", line 225, in main
run_test(cfg, model, args.distributed) #, model_name="model_final"
File "/home/jinyuda/scene_graph_benchmark/tools/train_sg_net.py", line 153, in run_test
save_predictions=cfg.TEST.SAVE_PREDICTIONS,
File "/home/jinyuda/scene_graph_benchmark/maskrcnn_benchmark/engine/inference.py", line 320, in inference
extra_args)
File "/home/jinyuda/scene_graph_benchmark/maskrcnn_benchmark/data/datasets/evaluation/init.py", line 31, in evaluate
return openimages_vrd_evaluation(args)
File "/home/jinyuda/scene_graph_benchmark/maskrcnn_benchmark/data/datasets/evaluation/openimages_vrd/init.py", line 9, in openimages_vrd_evaluation
force_relation=force_relation,
File "/home/jinyuda/scene_graph_benchmark/maskrcnn_benchmark/data/datasets/evaluation/openimages_vrd/openimages_vrd_eval.py", line 21, in do_openimages_vrd_evaluation
img_gt_dict, triplet_gt_dict, phrase_gt_dict = prepare_vrd_groundtruths(dataset)
File "/home/jinyuda/scene_graph_benchmark/maskrcnn_benchmark/data/datasets/evaluation/openimages_vrd/openimages_vrd_eval.py", line 182, in prepare_vrd_groundtruths
img_label = dataset.get_imagelabel(idx)
AttributeError: 'OpenImagesVRDTSVDataset' object has no attribute 'get_imagelabel'
"
I track the thread, and I found that OpenImagesVRDTSVDataset indeed has no function named "get_imagelabel" .
BTW, I used cuda11.3 and pytorch 1.7 to build the program.
this problem has confused me for several days, if you have any solution ,I will be really appreciated.
the information is as follow: " INFO:maskrcnn_benchmark.inference:Prepare ground truth annotations. Traceback (most recent call last): File "/home/jinyuda/.pycharm_helpers/pydev/pydevd.py", line 1483, in _exec pydev_imports.execfile(file, globals, locals) # execute the script File "/home/jinyuda/.pycharm_helpers/pydev/_pydev_imps/_pydev_execfile.py", line 18, in execfile exec(compile(contents+"\n", file, 'exec'), glob, loc) File "/home/jinyuda/scene_graph_benchmark/tools/train_sg_net.py", line 229, in
main()
File "/home/jinyuda/scene_graph_benchmark/tools/train_sg_net.py", line 225, in main
run_test(cfg, model, args.distributed) #, model_name="model_final"
File "/home/jinyuda/scene_graph_benchmark/tools/train_sg_net.py", line 153, in run_test
save_predictions=cfg.TEST.SAVE_PREDICTIONS,
File "/home/jinyuda/scene_graph_benchmark/maskrcnn_benchmark/engine/inference.py", line 320, in inference
extra_args)
File "/home/jinyuda/scene_graph_benchmark/maskrcnn_benchmark/data/datasets/evaluation/init.py", line 31, in evaluate
return openimages_vrd_evaluation(args)
File "/home/jinyuda/scene_graph_benchmark/maskrcnn_benchmark/data/datasets/evaluation/openimages_vrd/init.py", line 9, in openimages_vrd_evaluation
force_relation=force_relation,
File "/home/jinyuda/scene_graph_benchmark/maskrcnn_benchmark/data/datasets/evaluation/openimages_vrd/openimages_vrd_eval.py", line 21, in do_openimages_vrd_evaluation
img_gt_dict, triplet_gt_dict, phrase_gt_dict = prepare_vrd_groundtruths(dataset)
File "/home/jinyuda/scene_graph_benchmark/maskrcnn_benchmark/data/datasets/evaluation/openimages_vrd/openimages_vrd_eval.py", line 182, in prepare_vrd_groundtruths
img_label = dataset.get_imagelabel(idx)
AttributeError: 'OpenImagesVRDTSVDataset' object has no attribute 'get_imagelabel'
"
I track the thread, and I found that OpenImagesVRDTSVDataset indeed has no function named "get_imagelabel" . BTW, I used cuda11.3 and pytorch 1.7 to build the program. this problem has confused me for several days, if you have any solution ,I will be really appreciated.