zxytim
commented
7 years ago Hi World, you'd better submit the issue at https://github.com/argman/EAST but not this in this "fast-gmm" repo. Nonetheless, you can try playing around with weight decay, e.g., add 1e-6 weight decay while training. If not working, use more weight decay, like 1e-5.
On Sat, Sep 16, 2017 at 7:47 AM world notifications@github.com wrote:
hi, have you tried pvanet as basenetwork? I tried pvanet using caffe but encountered overfitting problem. my training sets is 950 images from icdar 2015 trainningsets( the other 50 images as validation sets) and 229 images from icdar 2013. model is trained by online data augmentation which includes scaling and rotations between ±30°. iou loss overfits a lot that when trainning iou descend to 0.25 validation iou loss still stays high at 0.7. I think I have confirmed everything so much that I can not solve this problem. please help me, Mr. tim!!!!!!. I have cost two month on this problem, all my thanks for you!!
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/zxytim/fast-gmm/issues/2, or mute the thread https://github.com/notifications/unsubscribe-auth/AA2QT_6ArxFgYPaxy3nglPcSs17eO3bZks5siwyDgaJpZM4PZn_v .
blankWorld
hi, have you tried pvanet as basenetwork? I tried pvanet using caffe but encountered overfitting problem. my training sets is 950 images from icdar 2015 trainningsets( the other 50 images as validation sets) and 229 images from icdar 2013. model is trained by online data augmentation which includes scaling and rotations between ±30°. iou loss overfits a lot that when trainning iou descend to 0.25 validation iou loss still stays high at 0.7. I think I have confirmed everything so much that I can not solve this problem. please help me, Mr. tim!!!!!!. I have cost two month on this problem, all my thanks for you!!