X-Git-Url: https://www.fleuret.org/cgi-bin/gitweb/gitweb.cgi?p=pysvrt.git;a=blobdiff_plain;f=cnn-svrt.py;h=338e145c52197e5f4649f031a83a88870473de12;hp=d0704fff48c85fca80c5723c20cb369c0600a013;hb=664f39333e0ae1ed2dccc6ec15e6c458dc8af935;hpb=4c77eebce3c3914a58c548c606d045efdae2284a

diff --git a/cnn-svrt.py b/cnn-svrt.py
index d0704ff..338e145 100755
--- a/cnn-svrt.py
+++ b/cnn-svrt.py
@@ -248,12 +248,13 @@ class DeepNet2(nn.Module):
 
     def __init__(self):
         super(DeepNet2, self).__init__()
+        self.nb_channels = 512
         self.conv1 = nn.Conv2d(  1,  32, kernel_size=7, stride=4, padding=3)
-        self.conv2 = nn.Conv2d( 32, 256, kernel_size=5, padding=2)
-        self.conv3 = nn.Conv2d(256, 256, kernel_size=3, padding=1)
-        self.conv4 = nn.Conv2d(256, 256, kernel_size=3, padding=1)
-        self.conv5 = nn.Conv2d(256, 256, kernel_size=3, padding=1)
-        self.fc1 = nn.Linear(4096, 512)
+        self.conv2 = nn.Conv2d( 32, self.nb_channels, kernel_size=5, padding=2)
+        self.conv3 = nn.Conv2d(self.nb_channels, self.nb_channels, kernel_size=3, padding=1)
+        self.conv4 = nn.Conv2d(self.nb_channels, self.nb_channels, kernel_size=3, padding=1)
+        self.conv5 = nn.Conv2d(self.nb_channels, self.nb_channels, kernel_size=3, padding=1)
+        self.fc1 = nn.Linear(16 * self.nb_channels, 512)
         self.fc2 = nn.Linear(512, 512)
         self.fc3 = nn.Linear(512, 2)
 
@@ -276,7 +277,7 @@ class DeepNet2(nn.Module):
         x = fn.max_pool2d(x, kernel_size=2)
         x = fn.relu(x)
 
-        x = x.view(-1, 4096)
+        x = x.view(-1, 16 * self.nb_channels)
 
         x = self.fc1(x)
         x = fn.relu(x)
@@ -539,7 +540,10 @@ for problem_number in map(int, args.problems.split(',')):
         else:
             validation_set = None
 
-        train_model(model, model_filename, train_set, validation_set, nb_epochs_done = nb_epochs_done)
+        train_model(model, model_filename,
+                    train_set, validation_set,
+                    nb_epochs_done = nb_epochs_done)
+
         log_string('saved_model ' + model_filename)
 
         nb_train_errors = nb_errors(model, train_set)