Is fitcensemble blocked when called from the MATLAB engine in Python?

I have a use case where I'm trying to call fitcensemble within a function that is called from the MATLAB engine within Python. It appears that the function cannot be found when called in this way. I make a call to trainAndSaveModel from within Python. This is a simplified version that I wrote to try and debug this issue.

function status = trainAndSaveModel(feature_matrix, label_matrix, savePath, learnerParams, trainingParams)
status = 1;
ver
which fitcensemble

produces this output:

----------------------------------------------------------------------------------------------------
MATLAB Version: 9.0.0.341360 (R2016a)
MATLAB License Number: ••••••
Operating System: Linux 4.4.0-75-generic #96-Ubuntu SMP Thu Apr 20 09:56:33 UTC 2017 x86_64
Java Version: Java 1.7.0_60-b19 with Oracle Corporation Java HotSpot(TM) 64-Bit Server VM mixed mode

----------------------------------------------------------------------------------------------------
MATLAB                                                Version 9.0         (R2016a)
Image Processing Toolbox                              Version 9.4         (R2016a)
Neural Network Toolbox                                Version 9.0         (R2016a)
Parallel Computing Toolbox                            Version 6.8         (R2016a)
Signal Processing Toolbox                             Version 7.2         (R2016a)
Statistics and Machine Learning Toolbox               Version 10.2        (R2016a)
'fitcensemble' not found.

Whereas if I just call which on fitcensemble from MATLAB directory it is able to find it:

which fitcensemble
/usr/local/MATLAB/R2016b/toolbox/stats/classreg/fitcensemble.m

I looked through some documentation and didn't see this as a limitation. Can anyone help me on this or do I have to make a system call to MATLAB to access fitcensemble?

Answer 1

[Walter Roberson]

Your invocation from python is getting a MATLAB R2016 A instance, but fitcensemble is on your MATLAB R2016 B path.