Summary

Sourcehttps://www.tensorflow.org/

License:  Apache-2.0

Path:  Part of the conda installation

Documentationhttps://www.tensorflow.org/guide/keras https://keras.io/

Keras is a high-level neural networks API, written in Python and capable of running on top of TensorFlow, CNTK, or Theano. It was developed with a focus on enabling fast experimentation. Being able to go from idea to result with the least possible delay is key to doing good research.Use Keras if you need a deep learning library that:

  • Allows for easy and fast prototyping (through user friendliness, modularity, and extensibility).
  • Supports both convolutional networks and recurrent networks, as well as combinations of the two.
  • Runs seamlessly on CPU and GPU.

Using  keras

keras became a part of Tensorflow, and can be used from the Tensorflow installation as well as "standalone". To initialize the environment use the module command:

# get some information about keras
[elx]% xwhich keras

 No executable named keras found 
 Documentation: https://confluence.desy.de/display/IS/keras
 URL:           https://www.tensorflow.org/
 Manual:        https://keras.io/
 License:       Apache-2.0

# Using the standard CPU based installation
[elx]% module load maxwell conda/3.6
[elx]% which python
/software/anaconda3/5.2/bin/python

[elx]% python
Python 3.6.7 | packaged by conda-forge | (default, Feb 28 2019, 09:07:38) 
[GCC 7.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.

>>> from __future__ import absolute_import, division, print_function, unicode_literals
>>> import tensorflow as tf
>>> from tensorflow.keras import layers

>>> print(tf.__version__)
1.10.0
>>> print(tf.keras.__version__)
2.1.6-tf
# ---

# Using the GPU based installation
[elx]% module load maxwell conda/3.6
[elx]% source activate Tensorflow-GPU

(Tensorflow-GPU) [elx]$ which python
/software/anaconda3/5.2/envs/Tensorflow-GPU/bin/python

(Tensorflow-GPU) [elx]$ python
>>> from __future__ import absolute_import, division, print_function, unicode_literals
>>> import tensorflow as tf
>>> from tensorflow.keras import layers

>>> print(tf.__version__)
1.12.0
>>> print(tf.keras.__version__)
2.1.6-tf

(Tensorflow-GPU) [elx]$ conda deactivate
[elx]$ 

# ---

# Using the standalone version (CPU only!)
[elx]% module load maxwell conda/3.6
[elx]% which python
/software/anaconda3/5.2/bin/python

[elx]% python
Python 3.6.7 | packaged by conda-forge | (default, Feb 28 2019, 09:07:38) 
[GCC 7.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import keras
Using TensorFlow backend.
>>> keras.__version__
'2.2.4'