Incorrect Shape in the python notebook predict

Question

Closed this issue 7 years ago · 3 comments

I got the training working fine with the script. When loading the model for prediction, I get the following error:

ValueError: Error when checking : expected input_1 to have 2 dimensions, but got array with shape (100, 1, 1)

For line:

generated_images = g.predict([noise, sampled_labels], verbose=0)

I'm on the git version of theano and latest of keras (using theano backend, of course).

Answer 1 · 2017-01-04T04:02:13.000Z

Hnn, I guess it's a typo since in mnist_acgan.py, we have sampled_labels = np.array([[i] * 10 for i in range(10)]).reshape(-1, 1)?

Answer 2 · 2017-01-05T06:16:30.000Z

Ah yes, I updates the script itself but not the ipynb. Feel like opening a PR?

Answer 3 · 2017-01-05T06:26:10.000Z

Sure. I'll send one ASAP.