Update to Chainer 2.0

lib/models/vgg.py

@@ -7,41 +7,40 @@
 
 class VGG(chainer.Chain):
 
-    """Input dimensions are (224, 224)."""
+    """Input dimensions are (n, 3, 224, 224)."""
 
     def __init__(self):
-        super().__init__(
-            conv1_1=L.Convolution2D(3, 64, 3, stride=1, pad=1),
-            conv1_2=L.Convolution2D(64, 64, 3, stride=1, pad=1),
-            conv2_1=L.Convolution2D(64, 128, 3, stride=1, pad=1),
-            conv2_2=L.Convolution2D(128, 128, 3, stride=1, pad=1),
-            conv3_1=L.Convolution2D(128, 256, 3, stride=1, pad=1),
-            conv3_2=L.Convolution2D(256, 256, 3, stride=1, pad=1),
-            conv3_3=L.Convolution2D(256, 256, 3, stride=1, pad=1),
-            conv4_1=L.Convolution2D(256, 512, 3, stride=1, pad=1),
-            conv4_2=L.Convolution2D(512, 512, 3, stride=1, pad=1),
-            conv4_3=L.Convolution2D(512, 512, 3, stride=1, pad=1),
-            conv5_1=L.Convolution2D(512, 512, 3, stride=1, pad=1),
-            conv5_2=L.Convolution2D(512, 512, 3, stride=1, pad=1),
-            conv5_3=L.Convolution2D(512, 512, 3, stride=1, pad=1),
-            fc6=L.Linear(25088, 4096),
-            fc7=L.Linear(4096, 4096),
-            fc8=L.Linear(4096, 1000)
-        )
+        super(VGG, self).__init__()
+        with self.init_scope():
+            self.conv1_1 = L.Convolution2D(3, 64, 3, stride=1, pad=1)
+            self.conv1_2 = L.Convolution2D(64, 64, 3, stride=1, pad=1)
+            self.conv2_1 = L.Convolution2D(64, 128, 3, stride=1, pad=1)
+            self.conv2_2 = L.Convolution2D(128, 128, 3, stride=1, pad=1)
+            self.conv3_1 = L.Convolution2D(128, 256, 3, stride=1, pad=1)
+            self.conv3_2 = L.Convolution2D(256, 256, 3, stride=1, pad=1)
+            self.conv3_3 = L.Convolution2D(256, 256, 3, stride=1, pad=1)
+            self.conv4_1 = L.Convolution2D(256, 512, 3, stride=1, pad=1)
+            self.conv4_2 = L.Convolution2D(512, 512, 3, stride=1, pad=1)
+            self.conv4_3 = L.Convolution2D(512, 512, 3, stride=1, pad=1)
+            self.conv5_1 = L.Convolution2D(512, 512, 3, stride=1, pad=1)
+            self.conv5_2 = L.Convolution2D(512, 512, 3, stride=1, pad=1)
+            self.conv5_3 = L.Convolution2D(512, 512, 3, stride=1, pad=1)
+            self.fc6 = L.Linear(25088, 4096)
+            self.fc7 = L.Linear(4096, 4096)
+            self.fc8 = L.Linear(4096, 1000)
 
+        # Keep track of the pooling indices inside each function instance
         self.conv_blocks = [
             [self.conv1_1, self.conv1_2],
             [self.conv2_1, self.conv2_2],
             [self.conv3_1, self.conv3_2, self.conv3_3],
             [self.conv4_1, self.conv4_2, self.conv4_3],
             [self.conv5_1, self.conv5_2, self.conv5_3]
         ]
-        self.deconv_blocks= []
-
-        # Keep track of the pooling indices inside each function instance
-        self.mps = [F.MaxPooling2D(2, 2, use_cudnn=False) for _ in self.conv_blocks]
+        self.deconv_blocks = []
+        self.mps = [F.MaxPooling2D(2, 2) for _ in self.conv_blocks]
 
-    def __call__(self, x, train=False):
+    def __call__(self, x):
 
         """Return a softmax probability distribution over predicted classes."""
 
@@ -50,8 +49,8 @@ def __call__(self, x, train=False):
         h = hs[-1]
 
         # Fully connected layers
-        h = F.dropout(F.relu(self.fc6(h)), train=train)
-        h = F.dropout(F.relu(self.fc7(h)), train=train)
+        h = F.dropout(F.relu(self.fc6(h)))
+        h = F.dropout(F.relu(self.fc7(h)))
         h = self.fc8(h)
 
         return F.softmax(h)
@@ -65,13 +64,15 @@ def feature_map_activations(self, x):
         pre_pooling_sizes = []
 
         h = x
-
         for conv_block, mp in zip(self.conv_blocks, self.mps):
             for conv in conv_block:
                 h = F.relu(conv(h))
 
             pre_pooling_sizes.append(h.data.shape[2:])
-            h = mp(h)
+
+            # Disable cuDNN, else pooling indices will not be stored
+            with chainer.using_config('use_cudnn', 'never'):
+                h = mp(h)
             hs.append(h)
 
         return hs, pre_pooling_sizes

visualize.py

@@ -1,10 +1,15 @@
+import matplotlib  # NOQA
+matplotlib.use('Agg')  # NOQA
+
 import argparse
 import math
 import os
 
+import chainer
 from chainer import cuda
 from chainer import serializers
 from chainer import Variable
+from chainer import links as L
 import numpy as np
 import matplotlib.pyplot as plt
 import matplotlib.image as mpimg
@@ -127,19 +132,23 @@ def save_to_ims(activations, dirname):
         save_to_ims(activations, dirname)
 
 
-def main(args):
+if __name__ == '__main__':
+    args = parse_args()
     model = VGG()
     serializers.load_hdf5(args.model_filename, model)
 
+    im = read_im(args.image_filename)
+
     if args.gpu >= 0:
         cuda.get_device(args.gpu).use()
         model.to_gpu()
+        im = cuda.to_gpu(im)
 
-    im = read_im(args.image_filename)
-
-    visualize(args.out_dirname, im, model)
+    with chainer.using_config('train', False):
+        pred = model(im)
 
+    pred = pred.data
+    if cuda.get_array_module(pred) != np:
+        pred = cuda.to_cpu(pred)
 
-if __name__ == '__main__':
-    args = parse_args()
-    main(args)
+    visualize(args.out_dirname, im, model)