Fix examples and zoo after PoolingCUDNN refactor.

Author: lucasb-eyer

DeepFried2/zoo/vgg16.py

@@ -8,22 +8,22 @@ def model(fully_conv=True):
     return df.Sequential(
         conv3(  3, 64), df.ReLU(),
         conv3( 64, 64), df.ReLU(),
-        df.SpatialMaxPoolingCUDNN((2,2)),
+        df.PoolingCUDNN((2,2)),
         conv3( 64,128), df.ReLU(),
         conv3(128,128), df.ReLU(),
-        df.SpatialMaxPoolingCUDNN((2,2)),
+        df.PoolingCUDNN((2,2)),
         conv3(128,256), df.ReLU(),
         conv3(256,256), df.ReLU(),
         conv3(256,256), df.ReLU(),
-        df.SpatialMaxPoolingCUDNN((2,2)),
+        df.PoolingCUDNN((2,2)),
         conv3(256,512), df.ReLU(),
         conv3(512,512), df.ReLU(),
         conv3(512,512), df.ReLU(),
-        df.SpatialMaxPoolingCUDNN((2,2)),
+        df.PoolingCUDNN((2,2)),
         conv3(512,512), df.ReLU(),
         conv3(512,512), df.ReLU(),
         conv3(512,512), df.ReLU(),
-        df.SpatialMaxPoolingCUDNN((2,2)),
+        df.PoolingCUDNN((2,2)),
         *_vgg.model_head(fully_conv)
     )
 

DeepFried2/zoo/vgg19.py

@@ -8,25 +8,25 @@ def model(fully_conv=True):
     return df.Sequential(
         conv3(  3, 64), df.ReLU(),
         conv3( 64, 64), df.ReLU(),
-        df.SpatialMaxPoolingCUDNN((2,2)),
+        df.PoolingCUDNN((2,2)),
         conv3( 64,128), df.ReLU(),
         conv3(128,128), df.ReLU(),
-        df.SpatialMaxPoolingCUDNN((2,2)),
+        df.PoolingCUDNN((2,2)),
         conv3(128,256), df.ReLU(),
         conv3(256,256), df.ReLU(),
         conv3(256,256), df.ReLU(),
         conv3(256,256), df.ReLU(),
-        df.SpatialMaxPoolingCUDNN((2,2)),
+        df.PoolingCUDNN((2,2)),
         conv3(256,512), df.ReLU(),
         conv3(512,512), df.ReLU(),
         conv3(512,512), df.ReLU(),
         conv3(512,512), df.ReLU(),
-        df.SpatialMaxPoolingCUDNN((2,2)),
+        df.PoolingCUDNN((2,2)),
         conv3(512,512), df.ReLU(),
         conv3(512,512), df.ReLU(),
         conv3(512,512), df.ReLU(),
         conv3(512,512), df.ReLU(),
-        df.SpatialMaxPoolingCUDNN((2,2)),
+        df.PoolingCUDNN((2,2)),
         *_vgg.model_head(fully_conv)
     )
 

examples/MNIST/model.py

@@ -23,12 +23,12 @@ def lenet_cudnn():
     model.add(df.SpatialConvolutionCUDNN(1, 32, (5,5), border='same', bias=False))
     model.add(df.BatchNormalization(32))
     model.add(df.ReLU())
-    model.add(df.SpatialMaxPoolingCUDNN((2,2)))
+    model.add(df.PoolingCUDNN((2,2)))
 
     model.add(df.SpatialConvolutionCUDNN(32, 64, (5,5), border='same', bias=False))
     model.add(df.BatchNormalization(64))
     model.add(df.ReLU())
-    model.add(df.SpatialMaxPoolingCUDNN((2,2)))
+    model.add(df.PoolingCUDNN((2,2)))
     model.add(df.Reshape(-1, 7*7*64))
 
     model.add(df.Linear(7*7*64, 100, bias=False))

examples/MultipleOutputs/model.py

@@ -9,13 +9,13 @@ def cnn(size, *head):
         df.BatchNormalization(32), df.ReLU(),
         df.SpatialConvolutionCUDNN(32, 32, (3,3), border='same', bias=False),
         df.BatchNormalization(32), df.ReLU(),
-        df.SpatialMaxPoolingCUDNN((2,2)),
+        df.PoolingCUDNN((2,2)),
 
         df.SpatialConvolutionCUDNN(32, 64, (3,3), border='same', bias=False),
         df.BatchNormalization(64), df.ReLU(),
         df.SpatialConvolutionCUDNN(64, 64, (3,3), border='same', bias=False),
         df.BatchNormalization(64), df.ReLU(),
-        df.SpatialMaxPoolingCUDNN((2,2)),
+        df.PoolingCUDNN((2,2)),
 
         df.Reshape(-1, 64*(size//4)**2),
 

examples/Optimizers/model.py

@@ -23,12 +23,12 @@ def lenet():
     model.add(df.SpatialConvolutionCUDNN(1, 32, (5,5), border=(2,2), with_bias=False))
     model.add(df.BatchNormalization(32))
     model.add(df.ReLU())
-    model.add(df.SpatialMaxPoolingCUDNN((2,2)))
+    model.add(df.PoolingCUDNN((2,2)))
 
     model.add(df.SpatialConvolutionCUDNN(32, 64, (5,5), border=(2,2), with_bias=False))
     model.add(df.BatchNormalization(64))
     model.add(df.ReLU())
-    model.add(df.SpatialMaxPoolingCUDNN((2,2)))
+    model.add(df.PoolingCUDNN((2,2)))
     model.add(df.Reshape(-1, 7*7*64))
 
     model.add(df.Linear(7*7*64, 100, with_bias=False))