tensorflow
Verwendung von transponierten Faltungsschichten

Verwendung von tf.nn.conv2d_transpose für beliebige Stapelgrößen und mit automatischer Ausgabeformberechnung.

Beispiel für die Berechnung der Ausgabeform und die Überwindung der Schwierigkeiten bei der Verwendung von tf.nn.conv2d_transpose mit unbekannter Stapelgröße (wenn input.get_shape () (?, H, W, C) oder (?, C, H, W) ist ).

def upconvolution (input, output_channel_size, filter_size_h, filter_size_w,
                   stride_h, stride_w, init_w, init_b, layer_name, 
                   dtype=tf.float32, data_format="NHWC", padding='VALID'):
    with tf.variable_scope(layer_name):
      #calculation of the output_shape:
      if data_format == "NHWC":
        input_channel_size = input.get_shape().as_list()[3]
        input_size_h = input.get_shape().as_list()[1]
        input_size_w = input.get_shape().as_list()[2]
        stride_shape = [1, stride_h, stride_w, 1]
        if padding == 'VALID':
          output_size_h = (input_size_h - 1)*stride_h + filter_size_h
          output_size_w = (input_size_w - 1)*stride_w + filter_size_w
        elif padding == 'SAME':
          output_size_h = (input_size_h - 1)*stride_h + 1
          output_size_w = (input_size_w - 1)*stride_w + 1
        else:
          raise ValueError("unknown padding")
        output_shape = tf.stack([tf.shape(input)[0], 
                                output_size_h, output_size_w, 
                                output_channel_size])
      elif data_format == "NCHW":
        input_channel_size = input.get_shape().as_list()[1]
        input_size_h = input.get_shape().as_list()[2]
        input_size_w = input.get_shape().as_list()[3]
        stride_shape = [1, 1, stride_h, stride_w]
        if padding == 'VALID':
          output_size_h = (input_size_h - 1)*stride_h + filter_size_h
          output_size_w = (input_size_w - 1)*stride_w + filter_size_w
        elif padding == 'SAME':
          output_size_h = (input_size_h - 1)*stride_h + 1
          output_size_w = (input_size_w - 1)*stride_w + 1
        else:
          raise ValueError("unknown padding")
        output_shape = tf.stack([tf.shape(input)[0], 
                                output_channel_size, 
                                output_size_h, output_size_w])
      else:
        raise ValueError("unknown data_format")

      #creating weights:
      shape = [filter_size_h, filter_size_w, 
               output_channel_size, input_channel_size]
      W_upconv = tf.get_variable("w", shape=shape, dtype=dtype,
                                 initializer=init_w)
      
      shape=[output_channel_size]
      b_upconv = tf.get_variable("b", shape=shape, dtype=dtype, 
                                 initializer=init_b)
      
      upconv = tf.nn.conv2d_transpose(input, W_upconv, output_shape, stride_shape,
                                      padding=padding,
                                      data_format=data_format)
      output = tf.nn.bias_add(upconv, b_upconv, data_format=data_format)
      
      #Now output.get_shape() is equal (?,?,?,?) which can become a problem in the 
      #next layers. This can be repaired by reshaping the tensor to its shape:
      output = tf.reshape(output, output_shape)
      #now the shape is back to (?, H, W, C) or (?, C, H, W)
      
      return output