I try to install, among others, the keras modules for use in the PythonCallers etc...
I installed them doing
$ sudo /opt/fme-engine-2019/fme python -m pip install --target=/opt/fmeserver/resources/engine/Plugins/Python keras keras_metrics tensorflow scipyand reorder the sys.path so all stuff in this Plugins directory comes before all other python modules. This normally works to override versions of packages and such.
When doing
import kerasthe engine running the workspace terminates instantly.
Now when I do
$ /opt/fme-engine-2019/fme python
>> import kerasI get a very long summary of the HDF5 configuration from which I pasted the interesting bits below...
Strangely I do not have the libhdf5-100 package installed on my system, and installing this (version 1.10) does not solve the problem.
Any idea's how to go about this? No wonder the engine craches when a python-command results in an enormous amount of text being spewed out to STD-OUT.
Warning! ***HDF5 library version mismatched error***
The HDF5 header files used to compile this application do not match the version used by the HDF5 library to which this application is linked
...
...
You should recompile the application or check your shared library relatedsettings such as 'LD_LIBRARY_PATH'.You can, at your own risk, disable this warning by setting the environmentvariable 'HDF5_DISABLE_VERSION_CHECK' to a value of '1'.Setting it to 2 or higher will suppress the warning messages totally.
Headers are 1.10.4, library is 1.8.19
SUMMARY OF THE HDF5 CONFIGURATION
=================================
General Information:
-------------------
HDF5 Version: 1.8.19
Configured on: Wed Jul 5 09:54:00 PDT 2017
Configured by: safe@cfl-linux
Configure mode: production
Host system: x86_64-unknown-linux-gnu
Uname information: Linux cfl-linux 4.4.0-83-generic #106-Ubuntu SMP Mon Jun 26 17:54:43 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux
Byte sex: little-endian
Libraries: static, shared
Installation point: /home/safe/code/fme-2018.0/3rd/redistributable/free_use/hdf5/hdf5-1.8.19/hdf5…...
.....
....
Bye...
Aborted (core dumped)
