Advertisement
Guest User

Untitled

a guest
Jun 15th, 2019
84
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 4.19 KB | None | 0 0
  1. def get_nst_model(weights_dict):
  2.  
  3. layers = {}
  4.  
  5. image_shape = (IMAGE_SIZE,IMAGE_SIZE,3)
  6. layers['input'] = tf.Variable(initial_value=tf.initializers.random_normal().__call__((1,)+image_shape),expected_shape=(1,)+image_shape,
  7. name='nst_output',dtype=tf.float32)
  8.  
  9. layers['conv1_1'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['input'],weights_dict['layer_1'][0],padding='SAME',strides=(1,1)),
  10. weights_dict['layer_1'][1]))
  11. layers['conv1_2'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['conv1_1'],weights_dict['layer_2'][0],padding='SAME',strides=(1,1)),
  12. weights_dict['layer_2'][1]))
  13. layers['pool1'] = tf.nn.avg_pool(layers['conv1_2'],ksize=(1,2,2,1),strides=(1,2,2,1),padding='VALID')
  14.  
  15. layers['conv2_1'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['pool1'],weights_dict['layer_4'][0],padding='SAME',strides=(1,1)),
  16. weights_dict['layer_4'][1]))
  17. layers['conv2_2'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['conv2_1'],weights_dict['layer_5'][0],padding='SAME',strides=(1,1)),
  18. weights_dict['layer_5'][1]))
  19. layers['pool2'] = tf.nn.avg_pool(layers['conv2_2'],ksize=(1,2,2,1),strides=(1,2,2,1),padding='VALID')
  20.  
  21.  
  22. layers['conv3_1'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['pool2'],weights_dict['layer_7'][0],padding='SAME',strides=(1,1)),
  23. weights_dict['layer_7'][1]))
  24. layers['conv3_2'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['conv3_1'],weights_dict['layer_8'][0],padding='SAME',strides=(1,1)),
  25. weights_dict['layer_8'][1]))
  26. layers['conv3_3'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['conv3_2'],weights_dict['layer_9'][0],padding='SAME',strides=(1,1)),
  27. weights_dict['layer_9'][1]))
  28. layers['conv3_4'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['conv3_3'],weights_dict['layer_10'][0],padding='SAME',strides=(1,1)),
  29. weights_dict['layer_10'][1]))
  30.  
  31. layers['pool3'] = tf.nn.avg_pool(layers['conv3_4'],ksize=(1,2,2,1),strides=(1,2,2,1),padding='VALID')
  32.  
  33.  
  34. layers['conv4_1'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['pool3'],weights_dict['layer_12'][0],padding='SAME',strides=(1,1)),
  35. weights_dict['layer_12'][1]))
  36. layers['conv4_2'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['conv4_1'],weights_dict['layer_13'][0],padding='SAME',strides=(1,1)),
  37. weights_dict['layer_13'][1]))
  38. layers['conv4_3'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['conv4_2'],weights_dict['layer_14'][0],padding='SAME',strides=(1,1)),
  39. weights_dict['layer_14'][1]))
  40. layers['conv4_4'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['conv4_3'],weights_dict['layer_15'][0],padding='SAME',strides=(1,1)),
  41. weights_dict['layer_15'][1]))
  42.  
  43. layers['pool4'] = tf.nn.avg_pool(layers['conv4_4'],ksize=(1,2,2,1),strides=(1,2,2,1),padding='VALID')
  44.  
  45.  
  46.  
  47. layers['conv5_1'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['pool4'],weights_dict['layer_17'][0],padding='SAME',strides=(1,1)),
  48. weights_dict['layer_17'][1]))
  49. layers['conv5_2'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['conv5_1'],weights_dict['layer_18'][0],padding='SAME',strides=(1,1)),
  50. weights_dict['layer_18'][1]))
  51. layers['conv5_3'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['conv5_2'],weights_dict['layer_19'][0],padding='SAME',strides=(1,1)),
  52. weights_dict['layer_19'][1]))
  53. layers['conv5_4'] = tf.nn.relu(tf.nn.bias_add(tf.nn.convolution(layers['conv5_3'],weights_dict['layer_20'][0],padding='SAME',strides=(1,1)),
  54. weights_dict['layer_20'][1]))
  55.  
  56. layers['pool5'] = tf.nn.avg_pool(layers['conv5_4'],ksize=(1,2,2,1),strides=(1,2,2,1),padding='VALID')
  57.  
  58. return layers
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement