py-R-FCN

py-R-FCN copied to clipboard

coco training problem

11

when i training use follow command, ./experiments/scripts/rfcn_end2end_ohem.sh 1 ResNet-101 coco it print below I0918 14:51:27.386899 22785 net.cpp:775] Ignoring source layer prob Solving... I0918 14:51:28.109838 22785 solver.cpp:228] Iteration 0, loss =...

2017hack

There is a small typo in README.md. Should read `official` rather than `offcial`. Semi-automated pull request generated by https://github.com/timgates42/meticulous/blob/master/docs/NOTE.md

timgates42

Message type "caffe.LayerParameter" has no field named "psroi_pooling_param". in test.....(demorfcn.py)

wanghangege

I open your given project code by pycharm and to run demo_rfcn but consequence with anitfsencoding: unable to load the file system codec ModuleNotFoundError: No module named 'encodings' fault .can...

13668379467

I0310 13:43:27.631085 15354 net.cpp:774] Copying source layer res5c I0310 13:43:27.631091 15354 net.cpp:774] Copying source layer res5c_relu I0310 13:43:27.631095 15354 net.cpp:771] Ignoring source layer pool5 I0310 13:43:27.631100 15354 net.cpp:771] Ignoring source...

newforrestgump001

It will got error if image has no box inside it: ``` im_size: (800.0, 1066.0) scale: 1.665624976158142 height, width: (50, 67) rpn: gt_boxes.shape (0, 5, 1, 1) rpn: gt_boxes []...

lucasjinreal

Fine Tuning

3

How to fine tune your model? I don't have sufficient data to retrain your model from scratch.I want to fine tune your model on my data which has only two...

mzahran001

In demo_rfcn.py, if cfg.TEST.HAS_RPN is set to False, then how do I pass value for blob['rois'] in the _get_blobs function in /py-R-FCN/lib/fast_rcnn/test.py ?

1

I understand that blobs['rois'] is being set to None. However if None is sent to _get_rois_blob (in the if branch), then another function '_project_im_rois' is giving this (AttributeError: 'NoneType' object...

Prithviraj7

I am trying to train my own data using rfcn_alt_opt_5stage_ohem and pascal_voc dataset. But I cannot find any parameters related to iterations. I can find ITERS=xxx in rfcn_end2end.sh and rfcn_end2end_ohem.sh....

wander1985

./lib/datasets/voc_eval.py BB = BB [sorted_ind, :] IndexError: too many indices

1

Hi @Orpine , I want to train my own dataset with one target, which is just a binary classification problem. I modify the prototxt in /models/pascal_voc/ResNet-50/rfcn_alt_opt_5step_ohem/ and change each cls-num...

fbi0817