jmiller656/discogan-tensorflow

An implementation of DiscoGAN in tensorflow

discogan-tensorflow's Issues

Interpreting the results

Hey,
Just wanted to confirm how to interpret the results. Does the image generated in generator a -> b , say img0.jpg mean that it is translating the image 000001.jpg from the celebA dataset from a female to male?

Well, in case of two custom datasets, say A and B, what does the generated image say img140.jpg signify?

Regards,
Raju

Roadmap

Hi @jmiller656 , I am playing around with DiscoGAN and thanks for letting me experimenting with it using your examples.

Are you looking forward to make this project more stable or are you planning this as a one-off project?

Error while running the code

Hello,
I am getting the following error while running:

File "main.py", line 5, in
g = utils.generator(a)
File "/Users/haosun/Raju/DiscoGAN-Tensorflow-master/utils.py", line 135, in generator
conv_t_1 = conv_layer_t(conv_4,[4,4,im_size,int(conv_4.get_shape()[-1])],[im_size],[tf.shape(x)[0],im_size/8,im_size/8,im_size],activation=lrelu,batch_norm=g_bn3,name="g_deconv_1",reuse=reuse)
File "/Users/haosun/Raju/DiscoGAN-Tensorflow-master/utils.py", line 75, in conv_layer_t
h = conv2d_t(x,w,shape)+b
File "/Users/haosun/Raju/DiscoGAN-Tensorflow-master/utils.py", line 29, in conv2d_t
return tf.nn.conv2d_transpose(x,w,shape,strides=[1,stride,stride,1],padding=padding)
File "/Users/haosun/Raju/anaconda/lib/python3.6/site-packages/tensorflow/python/ops/nn_ops.py", line 1024, in conv2d_transpose
output_shape_ = ops.convert_to_tensor(output_shape, name="output_shape")
File "/Users/haosun/Raju/anaconda/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 676, in convert_to_tensor
as_ref=False)
File "/Users/haosun/Raju/anaconda/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 741, in internal_convert_to_tensor
ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
File "/Users/haosun/Raju/anaconda/lib/python3.6/site-packages/tensorflow/python/ops/array_ops.py", line 919, in _autopacking_conversion_function
return _autopacking_helper(v, inferred_dtype, name or "packed")
File "/Users/haosun/Raju/anaconda/lib/python3.6/site-packages/tensorflow/python/ops/array_ops.py", line 881, in _autopacking_helper
constant_op.constant(elem, dtype=dtype, name=str(i)))
File "/Users/haosun/Raju/anaconda/lib/python3.6/site-packages/tensorflow/python/framework/constant_op.py", line 102, in constant
tensor_util.make_tensor_proto(value, dtype=dtype, shape=shape, verify_shape=verify_shape))
File "/Users/haosun/Raju/anaconda/lib/python3.6/site-packages/tensorflow/python/framework/tensor_util.py", line 374, in make_tensor_proto
_AssertCompatible(values, dtype)
File "/Users/haosun/Raju/anaconda/lib/python3.6/site-packages/tensorflow/python/framework/tensor_util.py", line 302, in _AssertCompatible
(dtype.name, repr(mismatch), type(mismatch).name))
TypeError: Expected int32, got 8.0 of type 'float' instead.

Could you help to provide a fix? Seems we need to typecast specifically