Printing a generator in python tensor flow

问题:

I am trying to follow the tensor flow tutorial as described in this link

I am trying to print the predicted result as described :

print ("Predicted %d, Label: %d" % (classifier.predict(test_data[0]), test_labels[0]))

But I am not able to print the result. I am getting the following error.

    print ("Predicted %d, Label: %d" % (classifier.predict(test_data[0]), test_labels[0]))
TypeError: %d format: a number is required, not generator

How do I print the generator in python.

I tried to write a loop and iterate over the elements it didn't work and I tried to use next to print the generator. That also didn't work. How do I print it ?

回答1:

This is how I solved it

new_samples = np.array([test_data[8]], dtype=float)

y = list(classifier.predict(new_samples, as_iterable=True))
print('Predictions: {}'.format(str(y)))

print ("Predicted %s, Label: %d" % (str(y), test_labels[8]))

回答2:

No tensorflow here, so let's mock up a generator and test it against your print expression

In [11]: def predict(a, b):
    ...:     for i in range(10):
    ...:         yield i, i*i
    ...:         

In [12]: print('a:%d, b:%d'%(predict(0, 0)))
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-12-29ec761936ef> in <module>()
----> 1 print('a:%d, b:%d'%(predict(0, 0)))

TypeError: %d format: a number is required, not generator

So far, so good: I have the same problem that you've experienced.

The problem is, of course, that what you get when you call a generator function are not values but a generator object...

You have to iterate on the generator objects, using whatever is returned from each iteration, e.g.,

In [13]: print('\n'.join('a:%d, b:%d'%(i,j) for i, j in predict(0,0)))
a:0, b:0
a:1, b:1
a:2, b:4
a:3, b:9
a:4, b:16
a:5, b:25
a:6, b:36
a:7, b:49
a:8, b:64
a:9, b:81

or, if you don't like one-liners,

In [14]: for i, j in predict(0, 0):
    ...:     print('a:%d, b:%d'%(i,j))
    ...:     
a:0, b:0
a:1, b:1
a:2, b:4
a:3, b:9
a:4, b:16
a:5, b:25
a:6, b:36
a:7, b:49
a:8, b:64
a:9, b:81

In other words, you have to explicitly consume what the generator is producing.

回答3:

From the documentation:

Runs inference to determine the class probability predictions. (deprecated arguments)

SOME ARGUMENTS ARE DEPRECATED. They will be removed after 2016-09-15. Instructions for updating: The default behavior of predict() is changing. The default value for as_iterable will change to True, and then the flag will be removed altogether. The behavior of this flag is described below.

Try:

classifier.predict(x=test_data[0])