Re: [theano-users] Using shared variable as inputs to theano function (values not being updated)

Tue, 23 May 2017 15:14:23 -0700

‎I have edited the code now to include input var and target var in the givens for the function, but it still doesn't work. Yes I want to define the model with input var and target var, which I now do with given, since shared variables cannot be used directly as input 
From: Frédéric Bastien
Sent: Wednesday, May 24, 2017 12:00 AM
Subject: Re: [theano-users] Using shared variable as inputs to theano function (values not being updated)

Hi,

You don't use input_var or target_var in your Theano function. So Theano ignore there value. Did you wanted to define the model with input_var and target_var instead of X and Y? If so, that could work by calling set_value().

Frédéric

On Thu, May 11, 2017 at 5:55 PM Tara <[email protected]> wrote:
I am trying to combine pymc3 with Theano for a simple recurrent neural network.However, when I complete training and change the input of the shared variables to the test set, the values are not updated in the graph even though the shared variables are updated.
Any ideas will be appreciated.
Here is the code :

# CREATE PYMC3 + THEANO IMPLEMENTATION OF A SIMPLE RECURRENT NETWORK
import timeit
start = timeit.default_timer()
import theano
import theano.tensor as T
import numpy as np
import pymc3 as pm
from scipy.stats import mode
theano.config.compute_test_value = 'ignore'
 
input_dim = 2
output_dim = 2
### PARAMETERS OF THE MODEL ###
hidden_dim = 64
learning_rate = 0.1
nb_epochs = 10
 
np.random.seed(0)
 
# Initialization /placeholder values
X = T.dtensor3('X')
Y = T.dtensor3('Y')
 
# begin by generating dataset so we have an array of lists
# ....
 
NUM_EXAMPLES = 1500
test_input = X_data[NUM_EXAMPLES:]
test_output = y_data[NUM_EXAMPLES:]
 
train_input = X_data[:NUM_EXAMPLES]
train_output = y_data[:NUM_EXAMPLES]
 
input_var = theano.shared(np.asarray(train_input).astype(np.float64), borrow = True)
target_var = theano.shared(np.asarray(train_output).astype(np.float64), borrow = True)
 
# Reference
# From paper :IMPROVING PERFORMANCE OF RECURRENT NEURAL NETWORK WITH RELU NONLINEARITY
def norm_positive_definite(r):
    A = np.dot(r, r.transpose())/hidden_dim
    values, vectors = np.linalg.eig(A)
    e = np.amax(values)
    

--

---
You received this message because you are subscribed to a topic in the Google Groups "theano-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/theano-users/_sxgPvgMeYo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to [email protected].
For more options, visit https://groups.google.com/d/optout.

--

---
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