Hi all,

I am using "newff" command to create my network.
During this I havr my trainig pattern as

P = { [p1] [p2] [p3] [p4] }
T = { [t1] [t2] [t3] [t4] }

So as you can see, they in sequential - concurrent format.
When the following command is invoked, train( net, P,T )
how does the training occur ?

Does it go {p1, t1} update weight and then
{p2,t2} update weight .....
{p4, t4} update weight
for every epoch

Or is it like

{p1 p2 p3 p4 } { t1 t2 t3 t4} (like in parallel) update weight for every epoch

Does anyone know ?

Kind Regards