change mil loss to use scores instead of softmax prob

losses.py

@@ -84,7 +84,7 @@ def multilabel_margin_loss_inner(y_,y_pred_,margin):
     return tf.reduce_mean(margin_violation)
 
 
-def mil_loss(scores, y, type='obj', epsilon=1e-5):
+def mil_loss_prob(scores, y, type='obj', epsilon=1e-5):
     if type=='obj':
         log_prob = tf.nn.log_softmax(scores)
     elif type=='atr':
@@ -94,6 +94,16 @@ def mil_loss(scores, y, type='obj', epsilon=1e-5):
     loss = -tf.reduce_sum(max_region_scores)/tf.maximum(tf.reduce_sum(y),epsilon)
     return loss
 
+def mil_loss(scores, y, type='obj', epsilon=1e-5):
+    if type=='obj':
+        log_prob = scores
+    elif type=='atr':
+        log_prob = tf.log(tf.maximum(epsilon, tf.nn.sigmoid(scores)))
+
+    max_region_scores = tf.minimum(tf.reduce_max(log_prob*y,0)-1.0,0.0)
+    loss = -tf.reduce_mean(max_region_scores)#/tf.maximum(tf.reduce_sum(y),epsilon)
+    return loss
+
 
 if __name__=='__main__':
     scores = tf.constant([[0.2, 0.3, 0.7],[0.8, 0.2, 0.9]])