Add description to card metadata

#1
by julien-c HF staff - opened
Files changed (1) hide show
  1. README.md +38 -4
README.md CHANGED
@@ -1,6 +1,6 @@
1
  ---
2
  title: BLEU
3
- emoji: 🤗
4
  colorFrom: blue
5
  colorTo: red
6
  sdk: gradio
@@ -8,10 +8,44 @@ sdk_version: 3.0.2
8
  app_file: app.py
9
  pinned: false
10
  tags:
11
- - evaluate
12
- - metric
13
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
14
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
15
  # Metric Card for BLEU
16
 
17
 
 
1
  ---
2
  title: BLEU
3
+ emoji: 🤗
4
  colorFrom: blue
5
  colorTo: red
6
  sdk: gradio
 
8
  app_file: app.py
9
  pinned: false
10
  tags:
11
+ - evaluate
12
+ - metric
13
+ description: >-
14
+ BLEU (bilingual evaluation understudy) is an algorithm for evaluating the
15
+ quality of text which has been machine-translated from one natural language to
16
+ another.
17
+
18
+ Quality is considered to be the correspondence between a machine's output and
19
+ that of a human: "the closer a machine translation is to a professional human
20
+ translation,
21
+
22
+ the better it is" – this is the central idea behind BLEU. BLEU was one of the
23
+ first metrics to claim a high correlation with human judgements of quality,
24
+ and
25
+
26
+ remains one of the most popular automated and inexpensive metrics.
27
+
28
+
29
+ Scores are calculated for individual translated segments—generally
30
+ sentences—by comparing them with a set of good quality reference translations.
31
+
32
+ Those scores are then averaged over the whole corpus to reach an estimate of
33
+ the translation's overall quality. Intelligibility or grammatical correctness
34
 
35
+ are not taken into account[citation needed].
36
+
37
+
38
+ BLEU's output is always a number between 0 and 1. This value indicates how
39
+ similar the candidate text is to the reference texts, with values closer to 1
40
+
41
+ representing more similar texts. Few human translations will attain a score of
42
+ 1, since this would indicate that the candidate is identical to one of the
43
+
44
+ reference translations. For this reason, it is not necessary to attain a score
45
+ of 1. Because there are more opportunities to match, adding additional
46
+
47
+ reference translations will increase the BLEU score.
48
+ ---
49
  # Metric Card for BLEU
50
 
51