Hello. (Sorry for my bad English. English is not my first language)

I am a theano beginner. 
now I am learning theano using https://deeplearning.net/tutorial/. 
LogisticResgression and MLP contents are quite understandable. But when I 
tried to using affNIST dataset (set of 40x40 image, 
http://www.cs.toronto.edu/~tijmen/affNIST/32x/originals/ ), I encountered 
error like this

 

<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>it
 
seems like 'Theano.tensor.dot' function error. (50000x784) dot (1600x500). 
It is nonsense. 
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>
 
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>I
 
changed mlp.py like this 
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>
  
  classifier = MLP( 
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>
  
      rng=rng, 
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>
  
      input=x, 
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>
  
      n_in=40 * 40, 
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>
  
      n_hidden=n_hidden, 
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>
  
      n_out=10 
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>
 
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>
  
  ) 
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>n_in
 
was modified. So the second operand was changed from 784x500 to 1600x500. 
but the first operand was not changed. I don't know why. any advice is 
maybe helpful.  thanks in advance.  
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>sorry
 
for my bad english 
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>
  
<https://lh3.googleusercontent.com/-lPrCX1dat5M/V5nKj1KvTPI/AAAAAAAAAhY/sNl1qjSQf0EZv0jydRxMLtbynr5orlZsgCLcB/s1600/err.png>

-- 

--- 
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.

Reply via email to