Public | Automated Build

Last pushed: an hour ago
Short Description
Extended from Jupyter Notebook Scientific Python Stack in order to add Keras.
Full Description



Extended from Jupyter Notebook Scientific Python Stack which contains :

  • Jupyter Notebook 5.0.x
  • Anaconda Python 3.x
  • pandas, matplotlib, scipy, seaborn, scikit-learn, scikit-image, sympy, ...

And much more, check the Jupyter Github for full reference and usage. Everything is pre-installed for Python 3.x.

Image build is triggered with each update of Jupyter Notebook Scientific Python Stack and install bleeding edge versions of Keras.

For advanced usage and options, read more at Jupyter Notebook Scientific Python Stack.

I use this image to work with neural networks, so I added the specific libraries :

  • Keras: a minimalist, highly modular neural networks library, written in Python and capable of running on top of either TensorFlow or Theano. It was developed with a focus on enabling fast experimentation.
  • TensorFlow: open source software library for numerical computation using data flow graphs, now the default backend for Keras.
  • HDF5 for Python: optional dependency of Keras used to save / load weights for neural networks.

Note : in this setting, TensorFlow will only CPU enabled only.

Basic Use

docker run -d -v /$(pwd)/:/home/jovyan/work \
           -p 8888:8888 gaarv/jupyter-keras --NotebookApp.token=''

This will allow to have the current working directory path mounted directly into the guest and with no password

Docker Pull Command
Source Repository