Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

the function of random_perspective in dataloaders.py causes labels4 to become empty #27

Open
Tristesse-stk opened this issue Nov 15, 2022 · 5 comments

Comments

@Tristesse-stk
Copy link

When I run the train.py, the labels4 and segments4 become empty after the function of random_perspective in dataloaders.py.

dataloader_random_before
dataloader_random_after

@RizwanMunawar
Copy link
Owner

@Tristesse-stk! Well, Can you share the img4, labels4 and segments4 values with print command, so that I can understand the issue in more better way.

@Tristesse-stk
Copy link
Author

@RizwanMunawar Ok.
img4:
test_result
(shape:(1024,1024,3))
label4:
array([[ 0, 208.87, 477.1, 225.2, 503.84],
[ 0, 743.2, 239.22, 751.28, 252.04],
[ 0, 208.87, 886.1, 225.2, 912.84],
[ 0, 730.02, 981.52, 735.54, 996.42]], dtype=float32)
segments4:
[0]:
0:array([ 220.43, 477.1], dtype=float32)
1:array([ 208.87, 503.84], dtype=float32)
2:array([ 211.59, 501.12], dtype=float32)
3:array([ 220.36, 503.84], dtype=float32)
4:array([ 223.43, 503.44], dtype=float32)
5:array([ 225.2, 478.6], dtype=float32)
[1]:
0:array([ 744.07, 239.22], dtype=float32)
1:array([ 743.2, 242.59], dtype=float32)
2:array([ 746.07, 252.04], dtype=float32)
3:array([ 747.76, 249.85], dtype=float32)
4:array([ 749.28, 250.01], dtype=float32)
5:array([ 749.08, 248.65], dtype=float32)
6:array([ 751.28, 245.75], dtype=float32)
[2]:
0:array([ 220.43, 886.1], dtype=float32)
1:array([ 208.87, 912.84], dtype=float32)
2:array([ 211.59, 910.12], dtype=float32)
3:array([ 220.36, 912.84], dtype=float32)
4:array([ 223.43, 912.44], dtype=float32)
5:array([ 225.2, 887.6], dtype=float32)
[3]:
0:array([ 734.32, 981.52], dtype=float32)
1:array([ 730.02, 987.43], dtype=float32)
2:array([ 730.25, 991.92], dtype=float32)
3:array([ 732.39, 994.51], dtype=float32)
4:array([ 732.5, 996.08], dtype=float32)
5:array([ 734.07, 996.42], dtype=float32)
6:array([ 735.54, 990.69], dtype=float32)

@Tristesse-stk
Copy link
Author

@RizwanMunawar Maybe it will more convenient for you if I give my datasets including annotations to you?
This is my first time to do instance segmentation project, maybe there is something wrong in my annotations.

@RizwanMunawar
Copy link
Owner

@Tristesse-stk. Well! did you used Roboflow for data labeling?
Sure, you can reach me through LinkedIn from mentioned link below, and we can discuss.
https://pk.linkedin.com/in/muhammadrizwanmunawar

@Tristesse-stk
Copy link
Author

@RizwanMunawar Yes, I use Roboflow for data labeling.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants