realscout/deep-learning-images

Automating image builds for deep learning based applications since 2016

deep-learning-images's Issues

Tensorflow configuration error

When I import tensorflow in the Python interpreter it is unable to load cuDNN. Here is the output:

>>> import tensorflow
I tensorflow/stream_executor/dso_loader.cc:111] successfully opened CUDA library libcublas.so.8.0 locally
I tensorflow/stream_executor/dso_loader.cc:105] Couldn't open CUDA library libcudnn.so.5. LD_LIBRARY_PATH: 
I tensorflow/stream_executor/cuda/cuda_dnn.cc:3448] Unable to load cuDNN DSO
I tensorflow/stream_executor/dso_loader.cc:111] successfully opened CUDA library libcufft.so.8.0 locally
I tensorflow/stream_executor/dso_loader.cc:111] successfully opened CUDA library libcuda.so.1 locally
I tensorflow/stream_executor/dso_loader.cc:111] successfully opened CUDA library libcurand.so.8.0 locally

upgrade cuDNN to v5

requires updates to caffe, probably a good thing anyway

More detail in README

Hi,

Thanks for sharing this. Would you please add more detailed steps in README to how use this. I have no idea what Packer is and where should I start.

ImportError when importing tensorflow

Hi guys,

Thanks for this repo. However, after creating both the base and tensorflow AMIs and launching a new instance I ran into an exception. When I ssh onto the box and try to import tensorflow into the Python interpreter, this error is raised:

ImportError: /usr/local/bin/../lib/libstdc++.so.6: version 'CXXABI_1.3.8' not found (required by /usr/local/lib/python3.5/site-packages/tensorflow/python/_pywrap_tensorflow.so)

Any ideas?

Base installation fails with error "NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver"

I'm getting an error while running the base installation:

packer build -only=amazon-ebs base.json

These are the last few lines of the output:

amazon-ebs: Running hooks in /etc/ca-certificates/update.d...

amazon-ebs:
amazon-ebs: done.
amazon-ebs: done.
amazon-ebs: /sbin/ldconfig.real: /usr/lib/nvidia-375/libEGL.so.1 is not a symbolic link
amazon-ebs:
amazon-ebs: /sbin/ldconfig.real: /usr/lib32/nvidia-375/libEGL.so.1 is not a symbolic link
amazon-ebs:
amazon-ebs:
amazon-ebs: WARNING: apt does not have a stable CLI interface. Use with caution in scripts.
amazon-ebs:
amazon-ebs: NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.
amazon-ebs:
 ==> amazon-ebs: Terminating the source AWS instance...
 ==> amazon-ebs: Cleaning up any extra volumes...
 ==> amazon-ebs: No volumes to clean up, skipping
 ==> amazon-ebs: Deleting temporary security group...
 ==> amazon-ebs: Deleting temporary keypair...
 Build 'amazon-ebs' errored: Script exited with non-zero exit status: 9

 ==> Some builds didn't complete successfully and had errors:
 --> amazon-ebs: Script exited with non-zero exit status: 9

 ==> Builds finished but no artifacts were created.