I 'm wondering, if this for loop is just not possible in theano, as it seems no one can give an answer to it? Well, I hope there is an answer.
Thanks again. Am Dienstag, 31. Januar 2017 19:05:13 UTC+1 schrieb [email protected]: > > Hello :), > > I want to calculate the derivatives of a function, where I want to realize > the function in a less memory requirements way, so I need a for loop in > theano. > > For the case, where P_temp (see below) is constant, I found something in > the internet: > > http://deeplearning.net/software/theano ... /scan.html > <http://deeplearning.net/software/theano/library/scan.html> > > But for the case here below, I did not found anything: > > > 1. P = T.zeros((M,M)) > 2. for i in range(0, N): > 3. P_temp = np.exp(SIGMA_trf[i , :]) * np.cos(MU[i , :]) > 4. P = P + P_temp > > Can you help me with this or show me some reference? > > Thanks! > -- --- 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.
