okwrtdsh / anaconda3 Goto Github PK

Running the container:
sudo nvidia-docker run -it okwrtdsh/anaconda3:keras-9.2-cudnn7 bash

"attaching" to container:
sudo docker exec -it container_name bash

Running python repl and importing tensorflow to check whether gpu is enabled in tensorflow:

from tensorflow.python.client import device_lib
print(device_lib.list_local_devices())

which prints:

2018-07-05 14:56:47.988072: I tensorflow/core/platform/cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
[name: "/device:CPU:0"
device_type: "CPU"
memory_limit: 268435456
locality {
}
incarnation: 3815053792300585653
]

That shows device_type as cpu. Can this be a bug?

I'm in love with your level-based approach for these Dockers! Great job and thanks for sharing!
However, what would you recommend as the favorable approach to enrich them with my own requirements/dependencies?

For example, if I'd like to run Keras on my GPU, with some new dependencies which are not currently present in your base image..
Would you recommend to create a new Dockerfile, build FROM keras-10.0-cudnn7 as base image, add my dependencies here?

Or, should I change your base 10.0-cudnn7 Dockerfile and insert my own dependencies, then build everything from scratch?

When running $ jupyter notebook --generate-config the file is "generated" according to the response in /root/.jupyter/ folder. I can navigate to that folder in a container terminal and see the file listed, but cannot open or access it. And when looking for that folder with any other option on my server, it just doesn't exist.