smapler error fixes

robotics-research-center · Jan 2, 2016 · ac59257 · ac59257
1 parent 00031e9
commit ac59257
Show file tree

Hide file tree

Showing 5 changed files with 30 additions and 18 deletions.
diff --git a/model.png → DCGAN.png b/model.png → DCGAN.png
diff --git a/README.md b/README.md
@@ -3,7 +3,7 @@ DCGAN in Tensorflow
 
 Tensorflow implementation of [Deep Convolutional Generative Adversarial Networks](http://arxiv.org/abs/1511.06434). The referenced torch code can be found [here](https://github.com/soumith/dcgan.torch).
 
-![alt tag](model.png)
+![alt tag](DCGAN.png)
 
 
 Prerequisites

diff --git a/model.py b/model.py
@@ -59,7 +59,7 @@ def build_model(self):
         self.z = tf.placeholder(tf.float32, [None, self.z_dim])
 
         self.image_ = self.generator(self.z)
-        self.sampler = self.generator(self.z)
+        self.sampler = self.sampler(self.z)
 
         self.D = self.discriminator(self.image)
         self.D_ = self.discriminator(self.image_, reuse=True)
@@ -207,36 +207,37 @@ def generator(self, z, y=None):
     def sampler(self, z, y=None):
         tf.get_variable_scope().reuse_variables()
 
-        if y:
+        if self.y_dim:
             yb = tf.reshape(y, [None, 1, 1, self.y_dim])
             z = tf.concat(1, [z, y])
 
-            h0 = tf.nn.relu(self.bn0(linear(z, self.gfc_dim, 's_h0_lin')))
+            h0 = tf.nn.relu(self.bn0(linear(z, self.gfc_dim, 'g_h0_lin')))
             h0 = tf.concat(1, [h0, y])
 
-            h1 = tf.nn.relu(self.g_bn1(linear(z, self.gf_dim*2*7*7, 's_h1_lin')))
+            h1 = tf.nn.relu(self.g_bn1(linear(z, self.gf_dim*2*7*7, 'g_h1_lin')))
             h1 = tf.reshape(h1, [None, 7, 7, self.gf_dim * 2])
             h1 = conv_cond_concat(h1, yb)
 
-            h2 = tf.nn.relu(self.bn2(deconv2d(h1, self.gf_dim, name='h2')))
+            h2 = tf.nn.relu(self.bn2(deconv2d(h1, self.gf_dim, name='g_h2')))
             h2 = conv_cond_concat(h2, yb)
 
-            return tf.nn.sigmoid(deconv2d(h2, self.c_dim, name='h3'))
+            return tf.nn.sigmoid(deconv2d(h2, self.c_dim, name='g_h3'))
         else:
-            h0 = tf.nn.relu(self.g_bn0(linear(z, self.gf_dim*8*4*4, 's_h0_lin'),
-                                       train=False))
-            h0 = tf.reshape(h1, [None, 4, 4, self.gf_dim * 8])
+            # project `z` and reshape
+            h0 = tf.reshape(linear(z, self.gf_dim*8*4*4, 'g_h0_lin'),
+                            [-1, 4, 4, self.gf_dim * 8])
+            h0 = tf.nn.relu(self.g_bn0(h0, train=False))
 
-            h1 = deconv2d(h0, [None, 8, 8, self.gf_dim*4], name='h1')
-            h1 = tf.relu(self.g_bn1(h1, train=False))
+            h1 = deconv2d(h0, [self.batch_size, 8, 8, self.gf_dim*4], name='g_h1')
+            h1 = tf.nn.relu(self.g_bn1(h1, train=False))
 
-            h2 = deconv2d(h1, [None, 16, 16, self.gf_dim*2], name='h2')
-            h2 = tf.relu(self.g_bn2(h2, train=False))
+            h2 = deconv2d(h1, [self.batch_size, 16, 16, self.gf_dim*2], name='g_h2')
+            h2 = tf.nn.relu(self.g_bn2(h2, train=False))
 
-            h3 = deconv2d(h2, [None, 16, 16, self.gf_dim*1], name='h3')
-            h3 = tf.relu(self.g_bn3(h3, train=False))
+            h3 = deconv2d(h2, [self.batch_size, 16, 16, self.gf_dim*1], name='g_h3')
+            h3 = tf.nn.relu(self.g_bn3(h3, train=False))
 
-            h4 = deconv2d(h3, [None, 64, 64, 3], name='h4')
+            h4 = deconv2d(h3, [None, 64, 64, 3], name='g_h4')
 
             return tf.nn.tanh(h4)
 

diff --git a/ops.py b/ops.py
@@ -54,7 +54,8 @@ def __call__(self, x, train=True):
             return tf.nn.batch_norm_with_global_normalization(x,
                                                               mean,
                                                               variance,
-                                                              self.beta, self.gamma,
+                                                              self.beta,
+                                                              self.gamma,
                                                               self.epsilon, True)
 
 def binary_cross_entropy_with_logits(logits, targets, name=None):

diff --git a/utils.py b/utils.py
@@ -15,6 +15,16 @@ def get_image(image_path):
 def imread(path):
     return scipy.misc.imread(path).astype(np.float)
 
+def imsave(images, size, path):
+    h, w = images.shape[1], images.shape[2]
+    img = np.zeros((h * size[0], w * size[1], 3))
+
+    for idx, image in enumerate(images):
+        i = idx % size[1]
+        j = idx / size[1]
+        img[j*h:j*h+h, i*w:i*w+w, :] = image
+    return scipy.misc.imsave(path, img)
+
 def center_crop(x, ph, pw=None):
     if pw is None:
         pw = ph