pva-faster-rcnn icon indicating copy to clipboard operation
pva-faster-rcnn copied to clipboard

Error: Caffe.LayerParameter" has no field named "smooth_l1_loss_param"

Open yzhang123 opened this issue 8 years ago • 6 comments

Hi I am trying to use tools/test_net.py on my own test data with the model models/pvanet/pva9.1/PVA9.1_ImgNet_COCO_VOC0712plus.caffemodel and models/pvanet/pva9.1/faster_rcnn_train_test_21cls.pt

However, the error_message is: W0503 15:56:22.701463 25382 _caffe.cpp:123] Use this instead (with the named "weights" parameter): W0503 15:56:22.701469 25382 _caffe.cpp:125] Net('models/pvanet/pva9.1/faster_rcnn_train_test_21cls.pt', 1, weights='models/pvanet/pva9.1/PVA9.1_ImgNet_COCO_VOC0712plus.caffemodel') [libprotobuf ERROR google/protobuf/text_format.cc:245] Error parsing text-format caffe.NetParameter: 6478:24: Message type "caffe.LayerParameter" has no field named "smooth_l1_loss_param". F0503 15:56:22.715694 25382 upgrade_proto.cpp:79] Check failed: ReadProtoFromTextFile(param_file, param) Failed to parse NetParameter file: models/pvanet/pva9.1/faster_rcnn_train_test_21cls.pt *** Check failure stack trace: *** Aborted (core dumped)

Thanks alot for your help @sanghoon

yzhang123 avatar May 03 '17 16:05 yzhang123

Ok i fixed it, my caffe version was not correctly built, so layers were missing

yzhang123 avatar May 03 '17 21:05 yzhang123

@yzhang123 hi,could you tell me how you fix this error in detail, I have the same error. Which caffe version is correct?

EchoCong avatar May 19 '17 01:05 EchoCong

@yzhang123 I am also facing the same issue. Could you please tell me which version of caffe you used?

shaileshvedula avatar Jan 21 '18 01:01 shaileshvedula

@yzhang123, I am facing the problem, may you help me?

JingchangHuang avatar Sep 03 '18 09:09 JingchangHuang

@sanghoom, I am facing the same problem, may you help me?

JingchangHuang avatar Sep 03 '18 09:09 JingchangHuang

which caffe version should be installed

Jackandrome avatar Jun 10 '19 14:06 Jackandrome