We're currently using IBM WSL version 1.2.3. Since it comes preinstalled with keras and tensorflow for R, we want to make use of that. However, it seems like the standard installation is not able to locate the right python environment to run the tensorflow backend, or the keras module for that matter.
Below is an example of a possible error. The python configuration seems to show the necessary keras module, but the error is still showing that it can not be found.
library(keras)
k = backend()
returns
Error: Python module keras was not found.
Detected Python configuration:
python: /opt/conda/bin/python
libpython: /opt/conda2/lib/libpython2.7.so
pythonhome: /opt/conda2:/opt/conda2
version: 2.7.11 |Anaconda custom (64-bit)| (default, Dec 6 2015, 18:08:32) [GCC 4.4.7 20120313 (Red Hat 4.4.7-1)]
numpy: /user-home/_global_/python-2.7/numpy
numpy_version: 1.16.2
keras: /user-home/_global_/python-2.7/keras
python versions found:
/opt/conda/bin/python
/usr/bin/python
/usr/bin/python3
Traceback:
1. keras::backend()
2. keras$backend
3. `$.python.builtin.module`(keras, backend)
4. py_resolve_module_proxy(x)
5. stop(message, call. = FALSE)
