I am also using the same code for saving my trained network and predict on
my new dataset. The code I am using is https://github.com/aseveryn/deep-qa.
The main part is run_nnet.py. I could save and load my network by the same
code in
https://github.com/lzhbrian/RBM-DBN-theano-DL4J/blob/master/src/theano/code/logistic_sgd.py
. but I continously get the same errors in this part,
predict_model = theano.function(inputs=[classifier.input],
outputs=classifier.y_pred)
actually I am not certain what I need to put in the inputs according to my
code. Which one is right?
inputs=[main.predict_prob_batch.batch_iterator],
outputs=test_nnet.layers[-1].y_pred)
inputs=[predict_prob_batch.batch_iterator],
outputs=test_nnet.layers[-1].y_pred)
inputs=[MiniBatchIteratorConstantBatchSize.dataset],
outputs=test_nnet.layers[-1].y_pred)
each of them I try I got the errors:
ImportError: No module named MiniBatchIteratorConstantBatchSize
or
NameError: global name 'predict_prob_batch' is not defined
I would really appreciate if you could help me.
I also used this command for running the code but still the errors.
python -c 'from run_nnet import predict; from sgd_trainer import
MiniBatchIteratorConstantBatchSize; from MiniBatchIteratorConstantBatchSize
import dataset; print predict()'
python -c 'from run_nnet import predict; from sgd_trainer import *; from
MiniBatchIteratorConstantBatchSize import dataset; print predict()'
Thank you.
On Friday, September 11, 2015 at 12:19:20 PM UTC-4, Lazaros Polymenakos
wrote:
>
> Thank you very much Daniel. The test commands you posted below, helped me
> trace the problem(s)
>
> For the record: there were several problems that had to be fixed:
>
> 1) in copying the logistic_sgd and modifying it to run on my data, the
> line
>
> self.input = input
>
> was commented out.
>
> 2) The pickle file was never saved properly because of the above.
>
> 3) when running the predict() function from a python command line,
> importing just the file that contains it (in this case logistic_sgd) did
> not seem to be enough. When I finally did: from logistic_reg import *
> (with the above problems fixed) it worked!
>
> Many thanks for the help,
>
> LP
>
>
>
> On Thursday, September 10, 2015 at 7:45:23 AM UTC-4, Daniel Renshaw wrote:
>>
>> The LogisticRegression class in the Theano sample definitely has an
>> attribute called `ìnput`. Have you created your own class called
>> LogisticRegression that is could be conflicting with the sample code's
>> version?
>>
>> To help test this, what happens if you run the following code while in
>> the current directory of wherever you have saved the logistic_sgd.py script?
>>
>> import theano
>> import theano.tensor as tt
>> import logistic_sgd
>> a = logistic_sgd.LogisticRegression(tt.matrix(name='x'), 10, 20)
>> print a.input.name
>>
>> This should not cause any errors and should simply print `x`.
>>
>> Daniel
>>
>>
>> On 9 September 2015 at 16:46, Lazaros Polymenakos <l.poly...@gmail.com>
>> wrote:
>>
>>> Hello Daniel,
>>>
>>> The classifier2 is a typo that I introduced when I tranferred the code
>>> to post it.
>>>
>>> it should read classifier.input.
>>>
>>>
>>> Here is what the error is pasted again:
>>>
>>> logistic_reg.predict()
>>> Traceback (most recent call last):
>>> File "<stdin>", line 1, in <module>
>>> File "logistic_reg.py", line 180, in predict
>>> inputs=[classifier.input],
>>> AttributeError: 'LogisticRegression' object has no attribute 'input'
>>>
>>>
>>> Sorry about the confusion.
>>>
>>> Many thanks.
>>>
>>> LP
>>>
>>>
>>> On Tuesday, September 8, 2015 at 8:52:38 PM UTC-4, Lazaros Polymenakos
>>> wrote:
>>>>
>>>> Hello,
>>>>
>>>> I am fairly new to THEANO. I am trying out the examples and adapting
>>>> them to run new experiments.
>>>>
>>>> In the documentation, there is a piece of code for the Logistic
>>>> Regression example that is called predict()
>>>> and it is used to load a trained model and test on a set of data.
>>>>
>>>> I append the modified code. The actual data loads and the sizes are
>>>> correct.
>>>>
>>>> I get the following error:
>>>>
>>>> Traceback (most recent call last):
>>>> File "<stdin>", line 1, in <module>
>>>> File "logistic_reg.py", line 180, in predict
>>>> inputs=[classifier.input],
>>>> AttributeError: 'LogisticRegression' object has no attribute 'input'
>>>>
>>>>
>>>> I would appreciate some insight on what could be wrong and how I could
>>>> fix it.
>>>>
>>>>
>>>> def predict():
>>>> """
>>>> An example of how to load a trained model and use it
>>>> to predict labels.
>>>> """
>>>>
>>>> # load the saved model
>>>> classifier = cPickle.load(open('best_model.pkl'))
>>>> print classifier.b.get_value().shape
>>>> # compile a predictor function
>>>> predict_model = theano.function(
>>>> inputs=[classifier.input],
>>>> outputs=classifier.y_pred)
>>>>
>>>> # We can test it on some examples from test test
>>>> # dataset='mnist.pkl.gz'
>>>> # datasets = load_data(dataset)
>>>> # test_set_x, test_set_y = datasets[2]
>>>> # test_set_x = test_set_x.get_value()
>>>> train_set_x = numpy.load('m_trainA.npy')
>>>> train_set_y = numpy.load('m_labels1.npy')
>>>> test_set_x = numpy.load('m_testA.npy')
>>>> test_set_y = numpy.load('m_labels2.npy')
>>>>
>>>> predicted_values = predict_model(test_set_x[:10])
>>>> print ("Predicted values for the first 10 examples in test set:")
>>>> print predicted_values
>>>>
>>>> Many thanks in advance
>>>>
>>>> --
>>>
>>> ---
>>> 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 theano-users...@googlegroups.com.
>>> 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 theano-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
