search

zju3dv / clean-pvnet

Code for "PVNet: Pixel-wise Voting Network for 6DoF Pose Estimation" CVPR 2019 oral
Apache License 2.0
401 stars 105 forks source link

2d projetions metric, ADD metric, and 5 cm 5 degree metrics are all reported during linemod training of object cat #285

Closed monajalal closed 9 months ago

monajalal commented 10 months ago

image

(clean-pvnet) mona@ada:~/clean-pvnet$ python train_net.py --cfg_file configs/linemod.yaml model mycat cls_type cat
loading annotations into memory...
Done (t=0.12s)
creating index...
index created!
loading annotations into memory...
Done (t=1.18s)
creating index...
index created!
loading annotations into memory...
Done (t=0.03s)
creating index...
index created!
eta: 0:04:25  epoch: 0  step: 20  vote_loss: 0.2364  seg_loss: 0.5301  loss: 0.7665  data: 0.1348  batch: 0.4433  lr: 0.001000  max_mem: 14401
eta: 0:03:55  epoch: 0  step: 40  vote_loss: 0.1862  seg_loss: 0.3189  loss: 0.5051  data: 0.0501  batch: 0.3721  lr: 0.001000  max_mem: 18549
eta: 0:03:46  epoch: 0  step: 60  vote_loss: 0.1580  seg_loss: 0.2212  loss: 0.3792  data: 0.0520  batch: 0.4021  lr: 0.001000  max_mem: 18549
eta: 0:03:31  epoch: 0  step: 80  vote_loss: 0.1490  seg_loss: 0.1568  loss: 0.3059  data: 0.0491  batch: 0.3585  lr: 0.001000  max_mem: 18580
eta: 0:03:20  epoch: 0  step: 100  vote_loss: 0.1372  seg_loss: 0.1202  loss: 0.2575  data: 0.0478  batch: 0.3600  lr: 0.001000  max_mem: 18580
eta: 0:03:09  epoch: 0  step: 120  vote_loss: 0.1241  seg_loss: 0.0944  loss: 0.2184  data: 0.0464  batch: 0.3478  lr: 0.001000  max_mem: 18580

MESSAGES

eta: 0:03:37  epoch: 24  step: 14912  vote_loss: 0.0023  seg_loss: 0.0025  loss: 0.0048  data: 0.0612  batch: 0.4735  lr: 0.000500  max_mem: 19654
eta: 0:03:29  epoch: 24  step: 14932  vote_loss: 0.0026  seg_loss: 0.0030  loss: 0.0055  data: 0.0555  batch: 0.4088  lr: 0.000500  max_mem: 19654
eta: 0:03:21  epoch: 24  step: 14952  vote_loss: 0.0029  seg_loss: 0.0026  loss: 0.0056  data: 0.0616  batch: 0.4694  lr: 0.000500  max_mem: 19654
eta: 0:03:13  epoch: 24  step: 14972  vote_loss: 0.0024  seg_loss: 0.0026  loss: 0.0050  data: 0.0590  batch: 0.4458  lr: 0.000500  max_mem: 19654
eta: 0:03:05  epoch: 24  step: 14992  vote_loss: 0.0022  seg_loss: 0.0024  loss: 0.0047  data: 0.0612  batch: 0.4679  lr: 0.000500  max_mem: 19654
eta: 0:02:57  epoch: 24  step: 15012  vote_loss: 0.0022  seg_loss: 0.0029  loss: 0.0051  data: 0.0541  batch: 0.4331  lr: 0.000500  max_mem: 19654
eta: 0:02:49  epoch: 24  step: 15032  vote_loss: 0.0024  seg_loss: 0.0025  loss: 0.0049  data: 0.0582  batch: 0.4600  lr: 0.000500  max_mem: 19654
eta: 0:02:40  epoch: 24  step: 15052  vote_loss: 0.0024  seg_loss: 0.0028  loss: 0.0052  data: 0.0549  batch: 0.4155  lr: 0.000500  max_mem: 19654
eta: 0:02:32  epoch: 24  step: 15072  vote_loss: 0.0023  seg_loss: 0.0027  loss: 0.0050  data: 0.0557  batch: 0.4374  lr: 0.000500  max_mem: 19654
eta: 0:02:24  epoch: 24  step: 15092  vote_loss: 0.0026  seg_loss: 0.0028  loss: 0.0054  data: 0.0566  batch: 0.4272  lr: 0.000500  max_mem: 19654
eta: 0:02:16  epoch: 24  step: 15112  vote_loss: 0.0025  seg_loss: 0.0029  loss: 0.0054  data: 0.0531  batch: 0.4269  lr: 0.000500  max_mem: 19654
eta: 0:02:08  epoch: 24  step: 15132  vote_loss: 0.0024  seg_loss: 0.0030  loss: 0.0054  data: 0.0569  batch: 0.4321  lr: 0.000500  max_mem: 19654
eta: 0:02:00  epoch: 24  step: 15152  vote_loss: 0.0026  seg_loss: 0.0030  loss: 0.0056  data: 0.0594  batch: 0.4632  lr: 0.000500  max_mem: 19654
eta: 0:01:52  epoch: 24  step: 15172  vote_loss: 0.0023  seg_loss: 0.0026  loss: 0.0050  data: 0.0595  batch: 0.4595  lr: 0.000500  max_mem: 19654
eta: 0:01:44  epoch: 24  step: 15192  vote_loss: 0.0021  seg_loss: 0.0024  loss: 0.0045  data: 0.0640  batch: 0.4714  lr: 0.000500  max_mem: 19654
eta: 0:01:36  epoch: 24  step: 15212  vote_loss: 0.0023  seg_loss: 0.0028  loss: 0.0051  data: 0.0537  batch: 0.4158  lr: 0.000500  max_mem: 19654
eta: 0:01:28  epoch: 24  step: 15232  vote_loss: 0.0024  seg_loss: 0.0030  loss: 0.0055  data: 0.0503  batch: 0.3662  lr: 0.000500  max_mem: 19654
eta: 0:01:20  epoch: 24  step: 15252  vote_loss: 0.0022  seg_loss: 0.0026  loss: 0.0048  data: 0.0580  batch: 0.4519  lr: 0.000500  max_mem: 19654
eta: 0:01:12  epoch: 24  step: 15272  vote_loss: 0.0023  seg_loss: 0.0029  loss: 0.0052  data: 0.0541  batch: 0.4020  lr: 0.000500  max_mem: 19654
eta: 0:01:03  epoch: 24  step: 15292  vote_loss: 0.0021  seg_loss: 0.0025  loss: 0.0046  data: 0.0623  batch: 0.4555  lr: 0.000500  max_mem: 19654
eta: 0:00:55  epoch: 24  step: 15312  vote_loss: 0.0022  seg_loss: 0.0029  loss: 0.0050  data: 0.0496  batch: 0.3811  lr: 0.000500  max_mem: 19654
eta: 0:00:47  epoch: 24  step: 15332  vote_loss: 0.0020  seg_loss: 0.0026  loss: 0.0046  data: 0.0579  batch: 0.4386  lr: 0.000500  max_mem: 19654
eta: 0:00:39  epoch: 24  step: 15352  vote_loss: 0.0023  seg_loss: 0.0028  loss: 0.0051  data: 0.0525  batch: 0.3774  lr: 0.000500  max_mem: 19654
eta: 0:00:31  epoch: 24  step: 15372  vote_loss: 0.0022  seg_loss: 0.0026  loss: 0.0048  data: 0.0551  batch: 0.4264  lr: 0.000500  max_mem: 19654
eta: 0:00:23  epoch: 24  step: 15392  vote_loss: 0.0020  seg_loss: 0.0025  loss: 0.0046  data: 0.0579  batch: 0.4064  lr: 0.000500  max_mem: 19654
eta: 0:00:15  epoch: 24  step: 15412  vote_loss: 0.0020  seg_loss: 0.0023  loss: 0.0043  data: 0.0632  batch: 0.4675  lr: 0.000500  max_mem: 19654
eta: 0:00:07  epoch: 24  step: 15432  vote_loss: 0.0022  seg_loss: 0.0026  loss: 0.0048  data: 0.0570  batch: 0.4358  lr: 0.000500  max_mem: 19654
eta: 0:00:00  epoch: 24  step: 15449  vote_loss: 0.0021  seg_loss: 0.0031  loss: 0.0052  data: 0.0450  batch: 0.3804  lr: 0.000500  max_mem: 19654
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1002/1002 [00:56<00:00, 17.81it/s]
['vote_loss: 0.0028', 'seg_loss: 0.0029', 'loss: 0.0057']
2d projections metric: 0.0
ADD metric: 0.0
5 cm 5 degree metric: 0.0
mask ap70: 0.9930139720558883
eta: 0:04:02  epoch: 25  step: 15470  vote_loss: 0.0021  seg_loss: 0.0027  loss: 0.0048  data: 0.1419  batch: 0.5896  lr: 0.000500  max_mem: 19654
eta: 0:03:54  epoch: 25  step: 15490  vote_loss: 0.0022  seg_loss: 0.0028  loss: 0.0050  data: 0.0557  batch: 0.4061  lr: 0.000500  max_mem: 19654
eta: 0:03:46  epoch: 25  step: 15510  vote_loss: 0.0022  seg_loss: 0.0023  loss: 0.0045  data: 0.0599  batch: 0.4621  lr: 0.000500  max_mem: 19654
eta: 0:03:38  epoch: 25  step: 15530  vote_loss: 0.0021  seg_loss: 0.0030  loss: 0.0052  data: 0.0514  batch: 0.4009  lr: 0.000500  max_mem: 19654
eta: 0:03:29  epoch: 25  step: 15550  vote_loss: 0.0021  seg_loss: 0.0025  loss: 0.0047  data: 0.0605  batch: 0.4500  lr: 0.000500  max_mem: 19654
eta: 0:03:21  epoch: 25  step: 15570  vote_loss: 0.0022  seg_loss: 0.0028  loss: 0.0051  data: 0.0525  batch: 0.4168  lr: 0.000500  max_mem: 19654
eta: 0:03:13  epoch: 25  step: 15590  vote_loss: 0.0023  seg_loss: 0.0028  loss: 0.0052  data: 0.0553  batch: 0.4369  lr: 0.000500  max_mem: 19654
eta: 0:03:05  epoch: 25  step: 15610  vote_loss: 0.0020  seg_loss: 0.0024  loss: 0.0045  data: 0.0635  batch: 0.4940  lr: 0.000500  max_mem: 19654
eta: 0:02:57  epoch: 25  step: 15630  vote_loss: 0.0021  seg_loss: 0.0026  loss: 0.0046  data: 0.0555  batch: 0.4178  lr: 0.000500  max_mem: 19654
eta: 0:02:49  epoch: 25  step: 15650  vote_loss: 0.0022  seg_loss: 0.0023  loss: 0.0045  data: 0.0632  batch: 0.4652  lr: 0.000500  max_mem: 19654
eta: 0:02:41  epoch: 25  step: 15670  vote_loss: 0.0021  seg_loss: 0.0025  loss: 0.0047  data: 0.0590  batch: 0.4581  lr: 0.000500  max_mem: 19654
eta: 0:02:33  epoch: 25  step: 15690  vote_loss: 0.0022  seg_loss: 0.0025  loss: 0.0047  data: 0.0579  batch: 0.4496  lr: 0.000500  max_mem: 19654
eta: 0:02:25  epoch: 25  step: 15710  vote_loss: 0.0020  seg_loss: 0.0027  loss: 0.0047  data: 0.0545  batch: 0.4111  lr: 0.000500  max_mem: 19654
eta: 0:02:17  epoch: 25  step: 15730  vote_loss: 0.0020  seg_loss: 0.0026  loss: 0.0046  data: 0.0553  batch: 0.4189  lr: 0.000500  max_mem: 19654

Any idea what could be wrong and how to fix this?

Vemokaz commented 7 months ago

Have you finished this problem yet?