I am trying to use tensorflow for implementing a dcgan and have run into this error:
ValueError: Shapes must be equal rank, but are 2 and 1
From merging shape 1 with other shapes. for 'generator/Reshape/packed' (op: 'Pack') with input shapes: [?,2048], [100,2048], [2048].
As far as I have gathered it indicates that my tensor shapes are different, but I cannot see what I need to change to fix this error. I believe the mistake hangs somewhere in between these methods:
First, I create a placeholder in a method using:
self.z = tf.placeholder(tf.float32, [None,self.z_dimension], name='z')
self.z_sum = tf.histogram_summary("z", self.z)
self.G = self.generator(self.z)
Then the last statement calls the generator method, this method uses reshape to change the tensor via:
self.z_ = linear(z,self.gen_dimension * 8 * sample_H16 * sample_W16, 'gen_h0_lin', with_w=True)
self.h0 = tf.reshape(self.z_,[-1, sample_H16, sample_W16,self.gen_dimension * 8])
h0 = tf.nn.relu(self.gen_batchnorm1(self.h0))
If it helps here is my linear method:
def linear(input_, output_size, scope=None, stddev=0.02, bias_start=0.0, with_w=False):
shape = input_.get_shape().as_list()
with tf.variable_scope(scope or "Linear"):
matrix = tf.get_variable("Matrix", [shape[1], output_size], tf.float32,tf.random_normal_initializer(stddev=stddev))
bias = tf.get_variable("bias", [output_size],initializer=tf.constant_initializer(bias_start))
if with_w:
return tf.matmul(input_, matrix) + bias, matrix, bias
else:
return tf.matmul(input_, matrix) + bias
EDIT:
I also use these placeholders:
self.inputs = tf.placeholder(tf.float32, shape=[self.batch_size] + image_dimension, name='real_images')
self.gen_inputs = tf.placeholder(tf.float32, shape=[self.sample_size] + image_dimension, name='sample_inputs')
inputs = self.inputs
sample_inputs = self.gen_inputs