CODEWITHSUNDEEP
pythontensorflowdeep-learningconvolutiondeconvolution
Sudip Das
Sudip Das

Reputation: 1208

How to make a de-convolution layer in tensorflow?

I have written a code for deconvolution layer,

def deconv2d(x, W,stride):
  x_shape = tf.shape(x)
  output_shape = tf.stack([x_shape[0], x_shape[1]*2, x_shape[2]*2, x_shape[3]//2])
  decon = tf.nn.conv2d_transpose(x, W, output_shape, strides=[1, stride, stride, 1], padding='SAME')
  layer_shape = get_layer_shape(decon)
  print('DECONV Shape : ', layer_shape)
  return  decon

I have called the above function like this way,

deconvolution1 = deconv2d(x=cnn_layer10, W=[2,2,512,1024], stride=2)

getting this error,

File "u-net.py", line 84, in obj.computation() File "u-net.py", line 41, in computation deconvolution1 = deconv2d(x=cnn_layer10, W=[2,2,512,1024], stride=2) File "/home/shuvo/u-net/architecture.py", line 35, in deconv2d decon = tf.nn.conv2d_transpose(x, W, output_shape, strides=[1, stride, stride, 1], padding='SAME') File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/nn_ops.py", line 1019, in conv2d_transpose if not value.get_shape()[axis].is_compatible_with(filter.get_shape()[3]):
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/tensor_shape.py", line 500, in getitem return self._dims[key] IndexError: list index out of range

My intention is to make a deconvolution layer where shape should be, [batch_size, 36,36,1024]=>[batch_size,72,72,512]. Please help me to fix this error,

Upvotes: 0

Views: 762

Answers (1)

Vijay Mariappan
Vijay Mariappan

Reputation: 17191

The input args filter to tf.nn.conv2d_transpose is the weights matrix itself and not just the size of the filter.

The modified code that fixes the above problem is shown below:

cnn_layer10 = tf.placeholder(tf.float32, (10, 36, 36, 1024))

def deconv2d(x, W,stride):
   x_shape = tf.shape(x)    
   weights = tf.Variable(tf.random_normal(W))
   output_shape = tf.stack([x_shape[0], x_shape[1]*2, x_shape[2]*2, x_shape[3]//2])
   decon = tf.nn.conv2d_transpose(x, weights, output_shape, strides=[1, stride, stride, 1], padding='SAME')

   return  decon

deconvolution1 = deconv2d(x=cnn_layer10, W=[2,2,512,1024], stride=2)

with tf.Session() as sess:
   sess.run(tf.global_variables_initializer())
   print('Decon Shape:',sess.run(deconvolution1, {cnn_layer10: np.random.random((10, 36,36,1024))}).shape)
   #Output
   #Decon Shape: (10, 72, 72, 512)

Note: Its better to use tf.layers.conv2d_transpose API, where the filters arg is the filter size and the weights initialization happens within.

Upvotes: 1

Related Questions