sarahwolf32/conditional-dcgan-for-mnist

A conditional DCGAN, in Tensorflow, for generating hand-written digits from the MNIST dataset.

conditional-dcgan-for-mnist's Issues

Shaping load_dataset

When running your code, I'm getting an attribute error:
AttributeError: 'tuple' object has no attribute 'shape'
for your dataset_loader class when trying to shape a loaded image:
num_batches = int(np.ceil(images.shape[0] / float(batch_size)))
in the dataset_loader.py

I did modify the input to take in mnist dataset from keras in the but I don't think that should be the root cause of this issue, it could be though. Any ideas? SO suggested casting it as a numpy.array(image) but that wasn't allowed.