VeloDC
commented
4 years ago Dear HuangLian126, Thank you for raising this issue, unfortunately we couldn't replicate the error on our side. I am still looking into it, meanwhile you could set logging_enabled=False in line 32 of "/home/hl/hl/oshot_detection-master/maskrcnn_benchmark/engine/oshot.py" (or comment out lines 98-99 of the same file) so that you can proceed with the evaluation
I have done the two steps 'Performing OSHOT pretraining' and 'Testing pretrained model', however get this error when running 'Performing OSHOT adaptation': ///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////// Traceback (most recent call last): File "tools/oshot_net.py", line 119, in
main()
File "tools/oshot_net.py", line 108, in main
cfg=cfg
File "/home/hl/hl/oshot_detection-master/maskrcnn_benchmark/engine/oshot.py", line 151, in oshot_inference
predictions = compute_on_dataset(model, data_loader, device, oshot_breakpoints, inference_timer, cfg)
File "/home/hl/hl/oshot_detection-master/maskrcnn_benchmark/engine/oshot.py", line 99, in compute_on_dataset
log_test_image(cfg, summarywriter, "detections{}_its".format(oshot_it+1), image_ids[0], images, output, image_name=image_name)
File "/home/hl/hl/oshot_detection-master/maskrcnn_benchmark/utils/log_image_bb.py", line 136, in log_test_image
log_image_and_bb(summary_writer, name, global_step, image, targets, image_name)
File "/home/hl/hl/oshot_detection-master/maskrcnn_benchmark/utils/log_image_bb.py", line 32, in log_image_and_bb
image = _overlay_boxes(image, targets)
File "/home/hl/hl/oshot_detection-master/maskrcnn_benchmark/utils/log_image_bb.py", line 98, in _overlay_boxes
colors = _compute_colors_for_labels(labels).tolist()
File "/home/hl/hl/oshot_detection-master/maskrcnn_benchmark/utils/log_image_bb.py", line 71, in _compute_colors_for_labels
cmcolors.append(np.array(cmap(lbl)[:3])*255)
File "/home/hl/anaconda3/envs/python37/lib/python3.7/site-packages/matplotlib/colors.py", line 561, in call
elif np.any(np.isnan(X)):
File "<__array_function__ internals>", line 6, in any
File "/home/hl/anaconda3/envs/python37/lib/python3.7/site-packages/numpy/core/fromnumeric.py", line 2330, in any
return _wrapreduction(a, np.logical_or, 'any', axis, None, out, keepdims=keepdims)
File "/home/hl/anaconda3/envs/python37/lib/python3.7/site-packages/numpy/core/fromnumeric.py", line 85, in _wrapreduction
return reduction(axis=axis, out=out, **passkwargs)
TypeError: any() received an invalid combination of arguments - got (out=NoneType, axis=NoneType, ), but expected one of: