Spatial pooler training and Temporal Memory

I have created a spatial pooler and temporal memory following the guide in BAMI. Can anyone comment on how best to train them? What I mean is should I train the spatial pooler for a set number of iterations or is it better to compute some sort of metric (if possible) to gauge how well the spatial pooler is settling and then run an while loop that runs until that metric is satisfied?

In a similar vein, since the spatial pooler generates SDRs for the temporal memory algorithm, how should the temporal memory algorithm be trained? I am thinking it is best to train the spatial pooler across a number of inputs to get it to stabilize and then to train the temporal memory but would there ever be a good case where one would want to train them both at the same time?

Thank you for any feedback.

1 Like