For example, as far as I know, GEMM fusion in an LSTM computation unit can speedup the model. But has it already been done by theano or I should do it in my implementation?
-- --- You received this message because you are subscribed to the Google Groups "theano-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
