how to train dynamic neural network on multiple sequences
in [Matlab]
|
From: Christian Vollmer on 22 Jan 2010 04:10 Hi, I'd like to train a layered recurrent network not only on a single sequence, but on multiple sequences. Suppose I have a set of input sequences (for example i1=[1 2 3] and i2=[1 2 2]). There is also a set of corresponding output sequences (for example t1=[2 3 4] and t2=[2 2 3]). One obvious way to train the network would be to concatenate all the input sequences to one big sequence i and all the target sequences to one big sequence t, then to convert them to cellarray via con2seq and then to train the network on that. But I was wondering if there is a way to train the network on all the sequences in a concurrent fashion. I know con2seq can convert multiple matrices to a batch of sequences (which is then a cell array with N rows for N sequences). I expected that the train function would then train the network on all the sequences concurrently, but the train function does only take a cell array with one row as input and target. Is there another way to do this? Thanks. -Christian
From: YEO IN_CHEOL on 25 Jan 2010 05:21 "Christian Vollmer" <christian.vollmer(a)gmx.net> wrote in message <hjbq28$al5$1(a)fred.mathworks.com>... > Hi, > > I'd like to train a layered recurrent network not only on a single sequence, but on multiple sequences. > Suppose I have a set of input sequences (for example i1=[1 2 3] and i2=[1 2 2]). There is also a set of corresponding output sequences (for example t1=[2 3 4] and t2=[2 2 3]). > One obvious way to train the network would be to concatenate all the input sequences to one big sequence i and all the target sequences to one big sequence t, then to convert them to cellarray via con2seq and then to train the network on that. > But I was wondering if there is a way to train the network on all the sequences in a concurrent fashion. I know con2seq can convert multiple matrices to a batch of sequences (which is then a cell array with N rows for N sequences). I expected that the train function would then train the network on all the sequences concurrently, but the train function does only take a cell array with one row as input and target. > > Is there another way to do this? Thanks. > > -Christian
|
Pages: 1 Prev: nmake not recognized in MATLAB 2009b Next: Convert a grayscale pixel to a rgb value |