Merge pull request #31 from maxisme/master

Fix TypeError and TensorFlow naming convention.

Author: lucasb-eyer

nets/resnet_v1_101.py

@@ -14,6 +14,6 @@ def endpoints(image, is_training):
         _, endpoints = resnet_v1_101(image, num_classes=None, is_training=is_training, global_pool=True)
 
     endpoints['model_output'] = endpoints['global_pool'] = tf.reduce_mean(
-        endpoints['resnet_v1_101/block4'], [1, 2], name='pool5', keep_dims=False)
+        endpoints['resnet_v1_101/block4'], [1, 2], name='pool5')
 
     return endpoints, 'resnet_v1_101'

nets/resnet_v1_50.py

@@ -14,6 +14,6 @@ def endpoints(image, is_training):
         _, endpoints = resnet_v1_50(image, num_classes=None, is_training=is_training, global_pool=True)
 
     endpoints['model_output'] = endpoints['global_pool'] = tf.reduce_mean(
-        endpoints['resnet_v1_50/block4'], [1, 2], name='pool5', keep_dims=False)
+        endpoints['resnet_v1_50/block4'], [1, 2], name='pool5')
 
     return endpoints, 'resnet_v1_50'

train.py

@@ -151,7 +151,7 @@ def sample_k_fids_for_pid(pid, all_fids, all_pids, batch_k):
     # create a padded list of indices which contain a multiple of the
     # original FID count such that all of them will be sampled equally likely.
     count = tf.shape(possible_fids)[0]
-    padded_count = tf.cast(tf.ceil(batch_k / count), tf.int32) * count
+    padded_count = tf.cast(tf.ceil(batch_k / tf.cast(count, tf.float32)), tf.int32) * count
     full_range = tf.mod(tf.range(padded_count), count)
 
     # Sampling is always performed by shuffling and taking the first k.