Message Boards Message Boards

Avoid error in ImageAugmentationLayer usage?

Posted 6 years ago

This code fails with "NetTrain::interr: An internal error occurred. Please contact Wolfram Research. " Any ideas?

lnet = NetChain[
   {
    ImageAugmentationLayer[{115, 115}, 
     "ReflectionProbabilities" -> {0.5, 0.5}],
    LinearLayer[1788],
    LogisticSigmoid,
    LinearLayer[1788],
    2,
    SoftmaxLayer[]},
   "Output" -> NetDecoder[{"Class", {1, 2}}],
   "Input" -> NetEncoder[{"Image", {125, 125}}]
   ] ;

tSet = Table[
  RandomChoice[{RegionImage[Annulus[], RasterSize -> 125] -> "A", 
    RegionImage[Circle[], RasterSize -> 125] -> "C"}], 10] 

NetTrain[lnet, tSet]
POSTED BY: Jeff Burns

This is definitely a bug. Thanks for reporting it! Will be fixed for 11.3.

Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard

Group Abstract Group Abstract