0

I am applying CNNs for semantic segmentation. I am hoping someone here can recommend. What I am doing now is that I scaled ground truth images. I have 5 classes that I scaled them to (0-1) in Data Layer based on this:

transform_param {
    scale: 0.00390625
  }

I am wondering whether I am wrong or right? Is this scaling value correct?

whenever I am not adding ` it is showing the following error:

[...]
I0510 23:21:28.086776  9072 solver.cpp:397]     Test net output #0: accuracy = 0
I0510 23:21:28.086812  9072 solver.cpp:397]     Test net output #1: loss = 1.9416 (* 1 = 1.9416 loss)
F0510 23:21:28.150539  9072 math_functions.cu:141] Check failed: status == CUBLAS_STATUS_SUCCESS (11 vs. 0)  CUBLAS_STATUS_MAPPING_ERROR
*** Check failure stack trace: ***
@     0x7fb9d4e7f5cd  google::LogMessage::Fail()
@     0x7fb9d4e81433  google::LogMessage::SendToLog()
@     0x7fb9d4e7f15b  google::LogMessage::Flush()
@     0x7fb9d4e81e1e  google::LogMessageFatal::~LogMessageFatal()
@     0x7fb9d56665ea  caffe::caffe_gpu_asum<>()
@     0x7fb9d5633a38  caffe::SoftmaxWithLossLayer<>::Forward_gpu()
@     0x7fb9d54bde41  caffe::Net<>::ForwardFromTo()
@     0x7fb9d54bdf47  caffe::Net<>::Forward()
@     0x7fb9d54e8d28  caffe::Solver<>::Step()
@     0x7fb9d54e98ca  caffe::Solver<>::Solve()
@           0x40acd4  train()
@           0x407418  main
@     0x7fb9d360f830  __libc_start_main
@           0x407ce9  _start
@              (nil)  (unknown)
S.EB
  • 1,966
  • 4
  • 29
  • 54
  • Are you scaling the labels???? – Shai May 06 '17 at 17:18
  • @Shai I have 5 classes, 0-4, and I added this transformation parameter to scale the labels between 0-1. Could you please guide? If I do not scale, it is giving me error `Check failed: status == CUBLAS STATUS_SUCCESS (11 vs. 0) CUBLAS STATUS MAPPING_ERROR`.Thanks – S.EB May 08 '17 at 02:10
  • what type of loss layer are you using? where do you get the error? please edit your question to reflect this new information. – Shai May 08 '17 at 08:23
  • 1
    If you have 5 classes, you should have them numbered 0-4. Do not scale the class designations. – Prune May 08 '17 at 16:09
  • @Shai Thanks for your comment. I am using `SoftmaxWithLoss` layer. – S.EB May 09 '17 at 15:30
  • @Prune Thanks for your help. I did not scale labels this time, but it still showing error. – S.EB May 09 '17 at 15:31

0 Answers0