fix: issues 1262, 1522 (#1655)

Co-authored-by: vasily <[email protected]>

deeppavlov/metrics/accuracy.py

@@ -154,6 +154,9 @@ def sets_accuracy(y_true: [list, np.ndarray], y_predicted: [list, np.ndarray]) -
 
     Returns:
         portion of samples with absolutely coincidental sets of predicted values
+
+    Alias:
+        sets_accuracy
     """
     examples_len = len(y_true)
     correct = sum([set(y1) == set(y2) for y1, y2 in zip(y_true, y_predicted)])

deeppavlov/metrics/fmeasure.py

@@ -27,6 +27,19 @@
 
 @register_metric('ner_f1')
 def ner_f1(y_true, y_predicted):
+    """
+    Calculates F1 measure for Named Entity Recognition task.
+
+    Args:
+        y_true: list of true values
+        y_predicted: list of predicted values
+
+    Returns:
+        F1 score
+
+    Alias:
+        ner_f1
+    """
     y_true = list(chain(*y_true))
     y_predicted = list(chain(*y_predicted))
     results = precision_recall_f1(y_true,
@@ -37,9 +50,23 @@ def ner_f1(y_true, y_predicted):
 
 
 @register_metric('ner_token_f1')
-def ner_token_f1(y_true, y_pred, print_results=False):
+def ner_token_f1(y_true, y_predicted, print_results=False):
+    """
+    Calculates F1 measure for Named Entity Recognition task without taking into account BIO or BIOES markup.
+
+    Args:
+        y_true: list of true values
+        y_predicted: list of predicted values
+        print_results: if True, then F1 score for each entity type is printed
+
+    Returns:
+        F1 score
+
+    Alias:
+        ner_f1
+    """
     y_true = list(chain(*y_true))
-    y_pred = list(chain(*y_pred))
+    y_pred = list(chain(*y_predicted))
 
     # Drop BIO or BIOES markup
     assert all(len(tag.split('-')) <= 2 for tag in y_true)
@@ -190,6 +217,9 @@ def round_f1(y_true, y_predicted):
 
     Returns:
         F1 score
+
+    Alias:
+        f1
     """
     try:
         predictions = [np.round(x) for x in y_predicted]
@@ -214,6 +244,9 @@ def round_f1_macro(y_true, y_predicted):
 
     Returns:
         F1 score
+
+    Alias:
+        f1_macro
     """
     try:
         predictions = [np.round(x) for x in y_predicted]
@@ -234,6 +267,9 @@ def round_f1_weighted(y_true, y_predicted):
 
     Returns:
         F1 score
+
+    Alias:
+        f1_weighted
     """
     try:
         predictions = [np.round(x) for x in y_predicted]

deeppavlov/metrics/log_loss.py

@@ -33,5 +33,8 @@ def sk_log_loss(y_true: Union[List[List[float]], List[List[int]], np.ndarray],
 
     Returns:
         Log loss
+
+    Alias:
+        log_loss
     """
     return log_loss(y_true, y_predicted)

deeppavlov/metrics/roc_auc_score.py

@@ -33,6 +33,9 @@ def roc_auc_score(y_true: Union[List[List[float]], List[List[int]], np.ndarray],
 
     Returns:
         Area Under the Curve (AUC) from prediction scores
+
+    Alias:
+        roc_auc
     """
     try:
         return sklearn.metrics.roc_auc_score(np.squeeze(np.array(y_true)),

docs/apiref/metrics.rst

@@ -13,6 +13,10 @@ Different Metric functions.
 
 .. autofunction:: deeppavlov.metrics.fmeasure.round_f1_weighted
 
+.. autofunction:: deeppavlov.metrics.fmeasure.ner_f1
+
+.. autofunction:: deeppavlov.metrics.fmeasure.ner_token_f1
+
 .. autofunction:: deeppavlov.metrics.log_loss.sk_log_loss
 
 .. autofunction:: deeppavlov.metrics.roc_auc_score.roc_auc_score