search

Yuliang-Liu / Curve-Text-Detector

This repository provides train＆test code, dataset, det.&rec. annotation, evaluation script, annotation tool, and ranking.
639 stars 155 forks source link

Failed to parse NetParameter file: ../models/ctd/test_ctd_tloc.prototxt #9

Closed earthat closed 5 years ago

earthat commented 6 years ago

HI

I am working on ubutu17.10 and installed caffe prebuilt bainaries for ubuntu>17.04. Everything was setup and when I run the demo.py in tools, I get this error

WARNING: Logging before InitGoogleLogging() is written to STDERR W0406 18:33:40.810593 4823 _caffe.cpp:139] DEPRECATION WARNING - deprecated use of Python interface W0406 18:33:40.810638 4823 _caffe.cpp:140] Use this instead (with the named "weights" parameter): W0406 18:33:40.810642 4823 _caffe.cpp:142] Net('../models/ctd/test_ctd_tloc.prototxt', 1, weights='../output/ctd_tloc.caffemodel') [libprotobuf ERROR google/protobuf/text_format.cc:298] Error parsing text-format caffe.NetParameter: 3767:25: Message type "caffe.LayerParameter" has no field named "psroi_pooling_param". F0406 18:33:40.812146 4823 upgrade_proto.cpp:88] Check failed: ReadProtoFromTextFile(param_file, param) Failed to parse NetParameter file: ../models/ctd/test_ctd_tloc.prototxt *** Check failure stack trace: *** Aborted (core dumped)

Please help me to resolve the issue.

With Regards

Yuliang-Liu commented 6 years ago

I haven't test my code on ubuntu17.10 (only tested on 14.04 and 16.04), and my code is based on a relative old caffe version with some old libraries (e.g., requiring protobuf 2). I'm not sure whether it can be run on a new caffe version.

The log showed "Message type "caffe.LayerParameter" has no field named "psroi_pooling_param"", which I guess is the main issue. I suggest you checking how psroipooling layer is defined and registered. Make sure it is included in caffe.proto file.

Regards, Yuliang Liu

earthat commented 6 years ago

yes, the parsing issue of this prototxt file is there. I tried by replacing the network in this file with googlenet and it was parsed. So I suppose it is included in caffe.proto file. How do you use or define the fastrcnn network. Is there any source form where you take it or idealize it as several fast-rcnn for caffe are available and I get just confused.

I also checked other resnet50 model from here, it was able to parse that also but not the model used in this code.

earthat commented 6 years ago

I don;t know exactly but I found that when this part of that prototxt file is removed then the error of psroi_pooling_param goes but another issue of cuda out of memory popup.

part of test_ctd_tloc_prototxt which causes issue problematic part of prototxt.txt

Please suggest any solution.

With Regards