This is the results we currently
provided as a proof of concept, but I am wondering if the above can be
obtained; possibly not in a graph, but if there is a work around or way of
re-producing this information it would be much appreciated.

I noticed on the 2015 page 205 - 215 on the non linear training - there seems to be a way of doing so. But some help in this situation would be much appreciated.

If you need any more details in relation to the model etc - please let me know ASAP