How to increase class label accuracy #4456
Unanswered
sarwarmurshed
asked this question in
Q&A
Replies: 1 comment
-
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
Hello,
I am using mask_rcnn_R_50_FPN_3x.yaml model on my private slap fingerprint dataset to segment the fingerprints and label them. In my dataset, There are 118K images and 10 classes. Classes are: left_index, left_middle, left_ring, left_little, right_index, right_middle, right_ring, right_little, left_thumb, right_thumb.
I am satisfied with the bounding box accuracy. However, not happy with the class label accuracy. The class label accuracy is only 94%. I am expecting to achieve 98% accuracy. I tried to train the model with different learning rates but had no luck.
Is there any way to improve class label accuracy? Any suggestion would be greatly appreciated.
Beta Was this translation helpful? Give feedback.
All reactions