update readme

tensorlayer · Jun 18, 2017 · cf51300 · cf51300
1 parent 7433735
commit cf51300
Show file tree

Hide file tree

Showing 3 changed files with 43 additions and 41 deletions.
diff --git a/README.md b/README.md
@@ -14,22 +14,22 @@ TensorFlow Implementation of ["Photo-Realistic Single Image Super-Resolution Usi
 </a>
 
 
-#### Results
+### Results
 
 <a href="http://tensorlayer.readthedocs.io">
 <div align="center">
 	<img src="img/SRGAN_Result2.png" width="80%" height="50%"/>
 </div>
 </a>
 
-#### Prepare Data and Pre-trained VGG
+### Prepare Data and Pre-trained VGG
 - In this experiment, we used images from [DIV2K - bicubic downscaling x4 competition](http://www.vision.ee.ethz.ch/ntire17/), you can also use your own data by setting your image folder in `config.py`
 
 - Download VGG model as [tutorial_vgg16.py](https://github.com/zsdonghao/tensorlayer/blob/master/example/tutorial_vgg16.py) show.
 
 
 
-#### Run
+### Run
 - Set your image folder in `config.py`.
 
 ```python

diff --git a/config.py b/config.py
@@ -15,7 +15,7 @@
     # config.TRAIN.decay_every_init = int(config.TRAIN.n_epoch_init / 2)
 
 ## adversarial learning (SRGAN)
-config.TRAIN.n_epoch = 1000
+config.TRAIN.n_epoch = 2000
 config.TRAIN.lr_decay = 0.1
 config.TRAIN.decay_every = int(config.TRAIN.n_epoch / 2)
 

diff --git a/model.py b/model.py
@@ -157,59 +157,61 @@ def SRGAN_d(input_images, is_train=True, reuse=False):
     b_init = None # tf.constant_initializer(value=0.0)
     gamma_init=tf.random_normal_initializer(1., 0.02)
     df_dim = 64
-    with tf.variable_scope("SRGAN_d2", reuse=reuse):
+    lrelu = lambda x: tl.act.lrelu(x, 0.2)
+    with tf.variable_scope("SRGAN_d", reuse=reuse):
         tl.layers.set_name_reuse(reuse)
-        net_in = InputLayer(input_images, name='d_input/images')
-        net_h0 = Conv2d(net_in, df_dim, (4, 4), (2, 2), act=lambda x: tl.act.lrelu(x, 0.2),
-                padding='SAME', W_init=w_init, name='d_h0/conv2d')
+        net_in = InputLayer(input_images, name='input/images')
+        net_h0 = Conv2d(net_in, df_dim, (4, 4), (2, 2), act=lrelu,
+                padding='SAME', W_init=w_init, name='h0/c')
 
         net_h1 = Conv2d(net_h0, df_dim*2, (4, 4), (2, 2), act=None,
-                padding='SAME', W_init=w_init, b_init=b_init, name='d_h1/conv2d')
-        net_h1 = BatchNormLayer(net_h1, act=lambda x: tl.act.lrelu(x, 0.2),
-                is_train=is_train, gamma_init=gamma_init, name='d_h1/batchnorm')
+                padding='SAME', W_init=w_init, b_init=b_init, name='h1/c')
+        net_h1 = BatchNormLayer(net_h1, act=lrelu, is_train=is_train,
+                gamma_init=gamma_init, name='h1/bn')
         net_h2 = Conv2d(net_h1, df_dim*4, (4, 4), (2, 2), act=None,
-                padding='SAME', W_init=w_init, b_init=b_init, name='d_h2/conv2d')
-        net_h2 = BatchNormLayer(net_h2, act=lambda x: tl.act.lrelu(x, 0.2),
-                is_train=is_train, gamma_init=gamma_init, name='d_h2/batchnorm')
+                padding='SAME', W_init=w_init, b_init=b_init, name='h2/c')
+        net_h2 = BatchNormLayer(net_h2, act=lrelu, is_train=is_train,
+                gamma_init=gamma_init, name='h2/bn')
         net_h3 = Conv2d(net_h2, df_dim*8, (4, 4), (2, 2), act=None,
-                padding='SAME', W_init=w_init, b_init=b_init, name='d_h3/conv2d')
-        net_h3 = BatchNormLayer(net_h3, act=lambda x: tl.act.lrelu(x, 0.2),
-                is_train=is_train, gamma_init=gamma_init, name='d_h3/batchnorm')
+                padding='SAME', W_init=w_init, b_init=b_init, name='h3/c')
+        net_h3 = BatchNormLayer(net_h3, act=lrelu, is_train=is_train,
+                gamma_init=gamma_init, name='h3/bn')
         net_h4 = Conv2d(net_h3, df_dim*16, (4, 4), (2, 2), act=None,
-                padding='SAME', W_init=w_init, b_init=b_init, name='d_h4/conv2d')
-        net_h4 = BatchNormLayer(net_h4, act=lambda x: tl.act.lrelu(x, 0.2),
-                is_train=is_train, gamma_init=gamma_init, name='d_h4/batchnorm')
+                padding='SAME', W_init=w_init, b_init=b_init, name='h4/c')
+        net_h4 = BatchNormLayer(net_h4, act=lrelu, is_train=is_train,
+                gamma_init=gamma_init, name='h4/bn')
         net_h5 = Conv2d(net_h4, df_dim*32, (4, 4), (2, 2), act=None,
-                padding='SAME', W_init=w_init, b_init=b_init, name='d_h5/conv2d')
-        net_h5 = BatchNormLayer(net_h5, act=lambda x: tl.act.lrelu(x, 0.2),
-                is_train=is_train, gamma_init=gamma_init, name='d_h5/batchnorm')
+                padding='SAME', W_init=w_init, b_init=b_init, name='h5/c')
+        net_h5 = BatchNormLayer(net_h5, act=lrelu, is_train=is_train,
+                gamma_init=gamma_init, name='h5/bn')
         net_h6 = Conv2d(net_h5, df_dim*16, (1, 1), (1, 1), act=None,
-                padding='SAME', W_init=w_init, b_init=b_init, name='d_h6/conv2d')
-        net_h6 = BatchNormLayer(net_h6, act=lambda x: tl.act.lrelu(x, 0.2),
-                is_train=is_train, gamma_init=gamma_init, name='d_h6/batchnorm')
+                padding='SAME', W_init=w_init, b_init=b_init, name='h6/c')
+        net_h6 = BatchNormLayer(net_h6, act=lrelu, is_train=is_train,
+                gamma_init=gamma_init, name='h6/bn')
         net_h7 = Conv2d(net_h6, df_dim*8, (1, 1), (1, 1), act=None,
-                padding='SAME', W_init=w_init, b_init=b_init, name='d_h7/conv2d')
-        net_h7 = BatchNormLayer(net_h7, #act=lambda x: tl.act.lrelu(x, 0.2),
-                is_train=is_train, gamma_init=gamma_init, name='d_h7/batchnorm')
+                padding='SAME', W_init=w_init, b_init=b_init, name='h7/c')
+        net_h7 = BatchNormLayer(net_h7, is_train=is_train,
+                gamma_init=gamma_init, name='h7/bn')
 
         net = Conv2d(net_h7, df_dim*2, (1, 1), (1, 1), act=None,
-                padding='SAME', W_init=w_init, b_init=b_init, name='d_h8_res/conv2d')
-        net = BatchNormLayer(net, act=lambda x: tl.act.lrelu(x, 0.2),
-                is_train=is_train, gamma_init=gamma_init, name='d_h8_res/batchnorm')
+                padding='SAME', W_init=w_init, b_init=b_init, name='res/c')
+        net = BatchNormLayer(net, act=lrelu, is_train=is_train,
+                gamma_init=gamma_init, name='res/bn')
         net = Conv2d(net, df_dim*2, (3, 3), (1, 1), act=None,
-                padding='SAME', W_init=w_init, b_init=b_init, name='d_h8_res/conv2d2')
-        net = BatchNormLayer(net, act=lambda x: tl.act.lrelu(x, 0.2),
-                is_train=is_train, gamma_init=gamma_init, name='d_h8_res/batchnorm2')
+                padding='SAME', W_init=w_init, b_init=b_init, name='res/c2')
+        net = BatchNormLayer(net, act=lrelu, is_train=is_train,
+                gamma_init=gamma_init, name='res/bn2')
         net = Conv2d(net, df_dim*8, (3, 3), (1, 1), act=None,
-                padding='SAME', W_init=w_init, b_init=b_init, name='d_h8_res/conv2d3')
-        net = BatchNormLayer(net, #act=lambda x: tl.act.lrelu(x, 0.2),
-                is_train=is_train, gamma_init=gamma_init, name='d_h8_res/batchnorm3')
-        net_h8 = ElementwiseLayer(layer=[net_h7, net], combine_fn=tf.add, name='d_h8/add')
+                padding='SAME', W_init=w_init, b_init=b_init, name='res/c3')
+        net = BatchNormLayer(net, is_train=is_train,
+                gamma_init=gamma_init, name='res/bn3')
+        net_h8 = ElementwiseLayer(layer=[net_h7, net],
+                combine_fn=tf.add, name='res/add')
         net_h8.outputs = tl.act.lrelu(net_h8.outputs, 0.2)
 
-        net_ho = FlattenLayer(net_h8, name='d_ho/flatten')
+        net_ho = FlattenLayer(net_h8, name='ho/flatten')
         net_ho = DenseLayer(net_ho, n_units=1, act=tf.identity,
-                W_init = w_init, name='d_ho/dense')
+                W_init = w_init, name='ho/dense')
         logits = net_ho.outputs
         net_ho.outputs = tf.nn.sigmoid(net_ho.outputs)