http://www.mathworks.com/matlabcentral/newsreader/view_thread/240913 Feed for thread: Re-Training net with different input size matrix( different sizes of input layer) en-us &copy;1994-2012 by MathWorks, Inc. webmaster@mathworks.com MATLAB Central Newsreader http://blogs.law.harvard.edu/tech/rss 60 http://www.mathworks.com/images/membrane_icon.gif Fri, 12 Dec 2008 14:10:06 -0500 http://www.mathworks.com/matlabcentral/newsreader/view_thread/240913#616632 zaheer ahmad Is it possible in Matlab to retrain a trained net with different input sized matrix( different sizes of input layers )? if yes then how to train the network using different input sized matrix i.e. i want to first train the network on 21x15 input matrix then want to train ( re-train ) the same network on 16x10 input matrix. The purpose of the process is to get a network which will be able to recognize different size of characters( to better generalize).....<br> I tried directly to pass new input matrix to 'train' but it produced error ( as i was expecting ).<br> one technique might be to pass it the weights and bias returned from the first training session but there will be a mismatch in the number of weights and bias because of change in number of input in input layer.... and i just got why the function 'train' trained on input size-1 matrix cant be trained directly on size-2 input matix....its because of number of weights too...am i correct?....then what can i do...how to retrain a network on differently sized input matrices...<br> I have read in some research papers people retraining net but i am not sure is it possible in matlab? <br> bye the way i am working on Urdu OCR ( Optical Character Recognition ), Urdu is national language of Pakistan.<br> thanks in advance<br> zaheer ahmad