Error occurred at Feature Extraction tutorial at the caffe website

1,376 views
Skip to first unread message

Eren Gölge

unread,
Oct 15, 2014, 5:53:18 AM10/15/14
to caffe...@googlegroups.com
As I try to run the given feature extraction example at the site, I got segmentation fault even-though I downloaded the bvlc_reference_caffenet.caffemodel file from the address given in the readme file. This is the error raised with model convention error.

E1015 12:42:21.547049  1503 extract_features.cpp:59] Using CPU
[libprotobuf ERROR google/protobuf/text_format.cc:245] Error parsing text-format caffe.NetParameter: 2:1: Invalid control characters encountered in text.
[libprotobuf ERROR google/protobuf/text_format.cc:245] Error parsing text-format caffe.NetParameter: 2:10: Invalid control characters encountered in text.
[libprotobuf ERROR google/protobuf/text_format.cc:245] Error parsing text-format caffe.NetParameter: 2:11: Message type "caffe.NetParameter" has no field named "CaffeNet".
F1015
12:42:21.904857  1503 upgrade_proto.cpp:623] Check failed: ReadProtoFromTextFile(param_file, param) Failed to parse NetParameter file: models/bvlc_reference_caffenet/bvlc_reference_caffenet.caffemodel
*** Check failure stack trace: ***
   
@     0x7fe1d481cdaa  (unknown)
   
@     0x7fe1d481cce4  (unknown)
   
@     0x7fe1d481c6e6  (unknown)
   
@     0x7fe1d481f687  (unknown)
   
@           0x4abb2e  caffe::ReadNetParamsFromTextFileOrDie()
   
@           0x461764  caffe::Net<>::Net()
   
@           0x415dc4  feature_extraction_pipeline<>()
   
@     0x7fe1d0e09ec5  (unknown)
   
@           0x412807  (unknown)
   
@              (nil)  (unknown)
Aborted (core dumped)


What would you suggest to solve that problem ?


DeepImage

unread,
May 15, 2015, 7:21:04 AM5/15/15
to caffe...@googlegroups.com
Did you find the solution to this problem? I am facing the same problem.

Carlos Treviño

unread,
Jun 17, 2015, 3:44:34 PM6/17/15
to caffe...@googlegroups.com
I'm also facing the same issue, any solution?

-Carlos

Eren Gölge

unread,
Jun 17, 2015, 4:34:56 PM6/17/15
to Carlos Treviño, caffe...@googlegroups.com
I have no solution still. Sorry. I used another library

--
You received this message because you are subscribed to a topic in the Google Groups "Caffe Users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/caffe-users/STI5fYSO4cs/unsubscribe.
To unsubscribe from this group and all its topics, send an email to caffe-users...@googlegroups.com.
To post to this group, send email to caffe...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/caffe-users/e03017db-efd8-40b1-a66e-277f20828c94%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.



--
​İyi günler dilerim
Eren Golge

8BİT Yapayzeka ve Görüntü İşleme Teknolojileri

Cyberpark Cyberplaza B Blok 2.Kat
No:B227 06800 Bilkent Çankaya/Ankara

TEL: 05052165749

npit

unread,
Jun 18, 2015, 7:04:44 AM6/18/15
to caffe...@googlegroups.com, trevino....@gmail.com
You must be either using a deprecated version of the network or an old caffe.cpp.
Can you link to the caffenet files you're using?

Kyle D

unread,
Nov 28, 2015, 1:49:02 AM11/28/15
to Caffe Users, trevino....@gmail.com
I also ran into this issue using this AMI (from the caffe website) and http://dl.caffe.berkeleyvision.org/bvlc_reference_caffenet.caffemodel model.

Vishnu Makkapati

unread,
Jan 15, 2016, 7:47:27 AM1/15/16
to Caffe Users
I run into the same error but while using GoogLeNet files. I converted the old files using upgrade_net_proto_binary and upgrade_net_proto_text after failing to load them. But the error still occurs.

Regards,
Vishnu

Mehmet Kerim Yucel

unread,
Mar 4, 2016, 2:31:53 PM3/4/16
to Caffe Users
I have managed to extract the features for the caffe official tutorial, though I am trying it for a different setting (on GoogLeNet trained for places, and dataset is chalearn activity recognition dataset). having the segmentation fault no matter what i do. Any updates regarding this problem?
Message has been deleted

Alex Quintero

unread,
Jul 20, 2016, 12:28:06 PM7/20/16
to Caffe Users
Check if you are passing the arguments in the right order. I was getting this error when I passed the caffemodel as the prototxt and vice versa. Switching them resolved the problem.
Reply all
Reply to author
Forward
0 new messages