BVLC / caffe

Caffe: a fast open framework for deep learning.

Home Page:http://caffe.berkeleyvision.org/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Error with mnist data normalisation using caffe container

ando2097 opened this issue · comments

Issue summary

Error when normalising mnist dataset after following instructions from: [NXP tutorial] (https://www.nxp.com/docs/en/application-note/AN12781.pdf)


root@6ced916972d5:/opt/caffe# build/tools/compute_image_mean -backend=lmdb examples/mnist/mnist_train_lmdb examples/mnist/
I0731 11:32:20.036314 1153 db_lmdb.cpp:35] Opened lmdb examples/mnist/mnist_train_lmdb
I0731 11:32:20.036592 1153 compute_image_mean.cpp:70] Starting iteration
I0731 11:32:20.041849 1153 compute_image_mean.cpp:95] Processed 10000 files.
I0731 11:32:20.047762 1153 compute_image_mean.cpp:95] Processed 20000 files.
I0731 11:32:20.054062 1153 compute_image_mean.cpp:95] Processed 30000 files.
I0731 11:32:20.060184 1153 compute_image_mean.cpp:95] Processed 40000 files.
I0731 11:32:20.065790 1153 compute_image_mean.cpp:95] Processed 50000 files.
I0731 11:32:20.071755 1153 compute_image_mean.cpp:95] Processed 60000 files.
I0731 11:32:20.071797 1153 compute_image_mean.cpp:108] Write to examples/mnist/
F0731 11:32:20.071946 1153 io.cpp:69] Check failed: proto.SerializeToOstream(&output)
*** Check failure stack trace: ***
@ 0x7f04629a15cd google::LogMessage::Fail()
@ 0x7f04629a3433 google::LogMessage::SendToLog()
@ 0x7f04629a115b google::LogMessage::Flush()
@ 0x7f04629a3e1e google::LogMessageFatal::~LogMessageFatal()
@ 0x7f0462f1b425 caffe::WriteProtoToBinaryFile()
@ 0x403052 main
@ 0x7f04618f3830 __libc_start_main
@ 0x403259 _start
@ (nil) (unknown)
Aborted


Steps to reproduce

follow instructions through to step 2 of caffe model training
(https://www.nxp.com/docs/en/application-note/AN12781.pdf)

Tried solutions

checking disk space (only 2% used)
checking permissions for examples/mnist/mnist_train_lmdb (had all necessary permissions)

System configuration

  • Operating system: Docker Image
  • Compiler: gcc version 5.4.0
  • CUDA version (if applicable):
  • CUDNN version (if applicable):
  • BLAS: N/A
  • Python version (if using pycaffe):
  • MATLAB version (if using matcaffe):

Issue checklist

  • [ *] read the guidelines and removed the first paragraph
  • [ *] written a short summary and detailed steps to reproduce
  • [* ] explained how solutions to related problems failed (tick if found none)
  • [* ] filled system configuration
  • [* ] attached relevant logs/config files (tick if not applicable)