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

Trying a bigger batch size #80

Open
bhushangawde opened this issue Jun 9, 2017 · 0 comments
Open

Trying a bigger batch size #80

bhushangawde opened this issue Jun 9, 2017 · 0 comments

Comments

@bhushangawde
Copy link

bhushangawde commented Jun 9, 2017

Hello @sanghoon
I am not able to fix this error!!!! I am trying a batch size of 4 instead of 1!
I know the problem is mismatch of rpn_cls_score , rpn_data and rpn_loss_cls!!
What should i do to change the layer configurations so that it will adapt to this!

OUTPUT-->

I0609 07:27:52.402818 18589 net.cpp:157] Top shape: 4 168 54 90 (3265920)
I0609 07:27:52.402824 18589 net.cpp:157] Top shape: 4 168 54 90 (3265920)
I0609 07:27:52.402829 18589 net.cpp:165] Memory required for data: 9355150164
I0609 07:27:52.402835 18589 layer_factory.hpp:77] Creating layer rpn_cls_score_reshape
I0609 07:27:52.402853 18589 net.cpp:100] Creating Layer rpn_cls_score_reshape
I0609 07:27:52.402859 18589 net.cpp:434] rpn_cls_score_reshape <- rpn_cls_score_rpn_cls_score_0_split_0
I0609 07:27:52.402868 18589 net.cpp:408] rpn_cls_score_reshape -> rpn_cls_score_reshape
I0609 07:27:52.402904 18589 net.cpp:150] Setting up rpn_cls_score_reshape
I0609 07:27:52.402911 18589 net.cpp:157] Top shape: 4 2 2268 90 (1632960)
I0609 07:27:52.402916 18589 net.cpp:165] Memory required for data: 9361682004
I0609 07:27:52.402921 18589 layer_factory.hpp:77] Creating layer rpn_cls_score_reshape_rpn_cls_score_reshape_0_split
I0609 07:27:52.402928 18589 net.cpp:100] Creating Layer rpn_cls_score_reshape_rpn_cls_score_reshape_0_split
I0609 07:27:52.402935 18589 net.cpp:434] rpn_cls_score_reshape_rpn_cls_score_reshape_0_split <- rpn_cls_score_reshape
I0609 07:27:52.402941 18589 net.cpp:408] rpn_cls_score_reshape_rpn_cls_score_reshape_0_split -> rpn_cls_score_reshape_rpn_cls_score_reshape_0_split_0
I0609 07:27:52.402951 18589 net.cpp:408] rpn_cls_score_reshape_rpn_cls_score_reshape_0_split -> rpn_cls_score_reshape_rpn_cls_score_reshape_0_split_1
I0609 07:27:52.402997 18589 net.cpp:150] Setting up rpn_cls_score_reshape_rpn_cls_score_reshape_0_split
I0609 07:27:52.403004 18589 net.cpp:157] Top shape: 4 2 2268 90 (1632960)
I0609 07:27:52.403009 18589 net.cpp:157] Top shape: 4 2 2268 90 (1632960)
I0609 07:27:52.403014 18589 net.cpp:165] Memory required for data: 9374745684
I0609 07:27:52.403019 18589 layer_factory.hpp:77] Creating layer rpn-data
I0609 07:27:52.403735 18589 net.cpp:100] Creating Layer rpn-data
I0609 07:27:52.403753 18589 net.cpp:434] rpn-data <- rpn_cls_score_rpn_cls_score_0_split_1
I0609 07:27:52.403760 18589 net.cpp:434] rpn-data <- gt_boxes_input-data_2_split_0
I0609 07:27:52.403767 18589 net.cpp:434] rpn-data <- im_info_input-data_1_split_0
I0609 07:27:52.403774 18589 net.cpp:434] rpn-data <- data_input-data_0_split_1
I0609 07:27:52.403781 18589 net.cpp:408] rpn-data -> rpn_labels
I0609 07:27:52.403791 18589 net.cpp:408] rpn-data -> rpn_bbox_targets
I0609 07:27:52.403800 18589 net.cpp:408] rpn-data -> rpn_bbox_inside_weights
I0609 07:27:52.403808 18589 net.cpp:408] rpn-data -> rpn_bbox_outside_weights
I0609 07:27:52.408412 18589 net.cpp:150] Setting up rpn-data
I0609 07:27:52.408437 18589 net.cpp:157] Top shape: 1 1 2268 90 (204120)
I0609 07:27:52.408444 18589 net.cpp:157] Top shape: 1 168 54 90 (816480)
I0609 07:27:52.408450 18589 net.cpp:157] Top shape: 1 168 54 90 (816480)
I0609 07:27:52.408457 18589 net.cpp:157] Top shape: 1 168 54 90 (816480)
I0609 07:27:52.408463 18589 net.cpp:165] Memory required for data: 9385359924
I0609 07:27:52.408468 18589 layer_factory.hpp:77] Creating layer rpn_loss_cls
I0609 07:27:52.408488 18589 net.cpp:100] Creating Layer rpn_loss_cls
I0609 07:27:52.408495 18589 net.cpp:434] rpn_loss_cls <- rpn_cls_score_reshape_rpn_cls_score_reshape_0_split_0
I0609 07:27:52.408504 18589 net.cpp:434] rpn_loss_cls <- rpn_labels
I0609 07:27:52.408512 18589 net.cpp:408] rpn_loss_cls -> rpn_loss_cls
I0609 07:27:52.408529 18589 layer_factory.hpp:77] Creating layer rpn_loss_cls
F0609 07:27:52.412192 18589 loss_layer.cpp:19] Check failed: bottom[0]->shape(0) == bottom[1]->shape(0) (4 vs. 1) The data and label should have the same first dimension.
*** Check failure stack trace: ***
Aborted

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

1 participant