description update

README.md

@@ -34,7 +34,7 @@ predictions = ["the cat is on the mat", "hello there"]
 references = ["the cat is on the mat", "hello there"]
 results = rougeraw.compute(predictions=predictions, references=references)
 print(results)
-{'rougeraw1_precision': 1.0, 'rougeraw1_recall': 1.0, 'rougeraw1_fmeasure': 1.0, 'rougeraw2_precision': 1.0, 'rougeraw2_recall': 1.0, 'rougeraw2_fmeasure': 1.0, 'rougerawl_precision': 1.0, 'rougerawl_recall': 1.0, 'rougerawl_fmeasure': 1.0}
+{'1_precision': 1.0, '1_recall': 1.0, '1_fmeasure': 1.0, '2_precision': 1.0, '2_recall': 1.0, '2_fmeasure': 1.0, 'l_precision': 1.0, 'l_recall': 1.0, 'l_fmeasure': 1.0}
 ```
 
 
@@ -43,22 +43,22 @@ predictions: list of predictions to evaluate. Each prediction should be a string
 references: list of reference for each prediction. Each reference should be a string with tokens separated by space
 
 ### Output Values
-- rougeraw1_precision
-- rougeraw1_recall
-- rougeraw1_fmeasure
-- rougeraw2_precision
-- rougeraw2_recall
-- rougeraw2_fmeasure
-- rougerawl_precision
-- rougerawl_recall
-- rougerawl_fmeasure
-
-Output Example(s):
-```python
-{'rougeraw1_precision': 1.0, 'rougeraw1_recall': 1.0, 'rougeraw1_fmeasure': 1.0, 'rougeraw2_precision': 1.0, 'rougeraw2_recall': 1.0, 'rougeraw2_fmeasure': 1.0, 'rougerawl_precision': 1.0, 'rougerawl_recall': 1.0, 'rougerawl_fmeasure': 1.0}
+This metric outputs a dictionary, containing the scores.
+
+There are precision, recall, F1 values for rougeraw-1, rougeraw-2 and rougeraw-l. By default the bootstrapped confidence intervals are calculated, meaning that for each metric there are low, mid , high values specifying the confidence interval.
+
+
+Key format: 
+```
+{1|2|l}_{low|mid|high}_{precision|recall|fmeasure}
+e.g.: 1_low_precision
 ```
 
-This metric outputs a dictionary, containing the scores.
+If aggregate is False the format is:
+```
+{1|2|l}_{precision|recall|fmeasure}
+e.g.: 1_precision
+```
 
 ## Citation(s)
 ```bibtex

rouge_raw.py

@@ -324,18 +324,20 @@ Args:
     select: (Optional) string. The name of the metric to return. One of: 'rougeraw1_precision', 'rougeraw1_recall', 'rougeraw1_fmeasure', 'rougeraw2_precision', 'rougeraw2_recall', 'rougeraw2_fmeasure', 'rougerawl_precision', 'rougerawl_recall', 'rougerawl_fmeasure'.
         If None, all metrics are returned as a dictionary.
 Returns:
-    1_precision
-    1_recall
-    1_fmeasure
-    2_precision
-    2_recall
-    2_fmeasure
-    l_precision
-    l_recall
-    l_fmeasure
+    This metric outputs a dictionary, containing the scores.
+    There are precision, recall, F1 values for rougeraw-1, rougeraw-2 and rougeraw-l. By default the bootstrapped confidence intervals are calculated, meaning that for each metric there are low, mid , high values specifying the confidence interval.
     
-    if aggregate is True there are also low, mid and high values for each metric. Thus, e.g.:
-        1_low_precision
+    Key format: 
+    ```
+    {1|2|l}_{low|mid|high}_{precision|recall|fmeasure}
+    e.g.: 1_low_precision
+    ```
+    
+    If aggregate is False the format is:
+    ```
+    {1|2|l}_{precision|recall|fmeasure}
+    e.g.: 1_precision
+    ```
 Examples:
     >>> rougeraw = evaluate.load('CZLC/rouge_raw')
     >>> predictions = ["the cat is on the mat", "hello there"]