Questions tagged [theano.scan]

Allows for looping in theano python library

Allows for looping in theano python library

36 questions
0
votes
1 answer

How to change inputs of the function given in theano.scan()?

I have confusion about theano.scan(). I have read the official documentation but i still feel my knowledge is limited. I want to change the inputs of the function given in theano.scan. For example, i have the following code. def…
Wasi Ahmad
  • 35,739
  • 32
  • 114
  • 161
0
votes
0 answers

how to print TensorVariable

I am a beginner of theano. I am studying it now. I'd like to print 'value' and 'shape' of TensorVariable while operating theano.function. When I used print fuction of python, print function ran before compiling theano function. So I learned using…
LKB
  • 457
  • 6
  • 16
0
votes
0 answers

how 'sequences' parameter works in theano scan function

In prior to get to the point, I apologize for my English sentences that can be rather awkward. Because English is not my first language. Now I am struggle with using theano.tensor.scan function properly. But, I don’t know how ‘sequences parameter’…
LKB
  • 457
  • 6
  • 16
0
votes
1 answer

Implementing Recurrence Equation (find Delta Coefficient similar to HTK) in theano

I would like to implement the following regression function with theano scan as an expression to compute delta coefficients. However, I can't figure out how to pass the previous inputs into the current step. where is a delta coefficient at time …
tolma
  • 141
  • 1
  • 10
0
votes
0 answers

How to realize different class weights in Theano

My problem is sentence classification, a two-class prediction problem: Does the sentence belong to class "1" or class "0"? In order to tackle the problem, I use a CNN. However, I want to force the model to punish errors on class "1" more, because it…
Alex
  • 778
  • 1
  • 8
  • 17
0
votes
1 answer

theano shared variable has wrong shape in scan function

I have a theano shared variable of shape (1, 500), but when passed to a scan function the shape turns out to be (1, 1, 500). Example code snippet is below. y_t1 = theano.shared(name='y_t1', value=np.zeros((1, 500), dtype=theano.config.floatX)) def…
blerud
  • 29
  • 1
  • 1
  • 5
1 2
3