deeplearning4j/deeplearning4j

Running a a basic CNN training fails on Google Colab!

Opened this issue · 0 comments

Issue Description

I am trying to Run a basic Deeplearking4j example, and building a similar version of the famous HotDog or not example. I am running the example on Google colab using jbang and IJava on GPU. Trying to run the code return an exception:

java.lang.NoClassDefFoundError: Could not initialize class org.bytedeco.javacv.OpenCVFrameConverter$ToMat
	at org.datavec.image.loader.NativeImageLoader.<init>(NativeImageLoader.java:59)
	at org.datavec.image.loader.NativeImageLoader.<init>(NativeImageLoader.java:116)
	at org.datavec.image.recordreader.BaseImageRecordReader.initialize(BaseImageRecordReader.java:131)
	at .(#245:1)

Hereis the link to the colab https://colab.research.google.com/drive/1m8BBTgkNMyfAnm4RaqTBERaht5qnNZ5D?usp=sharing
The same code works perfectly if I try to run it locally. Not sure If I missed some sort of dependency ?
Please describe our issue, along with:

  • expected behavior
    Training and saving the model
  • encountered behavior
    Failure and exception

Version Information

Please indicate relevant versions, including, if relevant:

  • Deeplearning4j version 1.0.0-M2.1
  • Platform information (OS, etc): Google Colab
  • CUDA version, if used
  • NVIDIA driver version, if in use

Additional Information

Where applicable, please also provide:

  • Full log or exception stack trace (ideally in a Gist: gist.github.com)
  • pom.xml file or similar (also in a Gist)

Contributing

If you'd like to help us fix the issue by contributing some code, but would
like guidance or help in doing so, please mention it!