fixed bug in metrics

metrics.py

@@ -149,13 +149,18 @@ def location_sensitive_detection(pred, true, n_frames=100, spatial_threshold=2.,
     for frame in range(n_frames):
         t = frames[frame]['t']  #all true events for frame i
         p = frames[frame]['p']  #all predicted events for frame i
-        matched = 0           #counts the matching events
-
-        if len(t) == 0:         #if there are PREDICTED but not TRUE events
-            FP += len(p)        #all predicted are false positive
-        elif len(p) == 0:       #if there are TRUE but not PREDICTED events
-            FN += len(t)        #all predicted are false negative
 
+        num_true_items = len(t)
+        num_pred_items = len(p)
+        matched = 0           #counts the matching events
+        match_ids = []       #all pred ids that matched
+
+        if num_true_items == 0:         #if there are PREDICTED but not TRUE events
+            FP += num_pred_items        #all predicted are false positive
+        elif num_pred_items == 0:       #if there are TRUE but not PREDICTED events
+            FN += num_true_items        #all predicted are false negative
+        elif num_true_items == 0 and num_pred_items == 0:  #if no true and no predicted, just do nothing
+            pass
         else:
             for i_t in range(len(t)):           #iterate all true events
                 match = False       #flag for matching events
@@ -168,18 +173,21 @@ def location_sensitive_detection(pred, true, n_frames=100, spatial_threshold=2.,
                     spat_error = np.linalg.norm(true_coord-pred_coord)  #cartesian distance between spatial coords
                     if true_class == pred_class and spat_error < spatial_threshold:  #if predicton is correct (same label + not exceeding spatial error threshold)
                         match = True
+                        match_ids.append(i_p)   #append to matched ids (to cound eventual duplicates)
                 if match:
                     matched += 1    #for each true event, match only once comparing all predicted events
 
-        num_true_items = len(t)
-        num_pred_items = len(p)
-        fn =  num_true_items - matched
-        fp = num_pred_items - matched
+            unique_ids = np.unique(match_ids)  #remove duplicates from matches ids list
+            duplicates = len(match_ids) - len(unique_ids)  #compute number of duplicates
+            matched = matched - duplicates
+            fn =  num_true_items - matched
+            fp = num_pred_items - matched
+
+            #add to counts
+            TP += matched          #number of matches are directly true positives
+            FN += fn
+            FP += fp
 
-        #add to counts
-        TP += matched          #number of matches are directly true positives
-        FN += fn
-        FP += fp
 
     precision = TP / (TP + FP + sys.float_info.epsilon)
     recall = TP / (TP + FN + sys.float_info.epsilon)