NilanE commited on
Commit
7e8620a
1 Parent(s): 91ecce1

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +21 -0
README.md CHANGED
@@ -1,3 +1,24 @@
1
  ---
2
  license: apache-2.0
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: apache-2.0
3
  ---
4
+
5
+ # Dataset details
6
+ Each entry in this dataset is a sentence-aligned Japanese web novel chapter and English fan translation.
7
+ The intended use-case is for document translation tasks.
8
+
9
+ # Dataset format
10
+ ```json
11
+ {
12
+ 'src' : 'JAPANESE CHAPTER'
13
+ 'trg' : 'ENGLISH TRANSLATION'
14
+ 'meta' : {
15
+ "source": 'SAME ACROSS ALL ENTRIES',
16
+ "series": 'NAME OF WEB NOVEL SERIES',
17
+ "missed_lines": 'NUMBER OF LINES THAT WERE AT THE SAME INDEX BUT NOT DETECTED AS BEING TRANSLATIONS OF EACH OTHER',
18
+ "inserted_lines_src": 'NUMBER OF LINES IN THE JAPANESE TEXT THAT DID NOT HAVE A MATCHING TRANSLATION BUT ARE BUFFERED BY TRANSLATED LINES',
19
+ "inserted_lines_trg": 'SAME AS ABOVE BUT FOR ENGLISH',
20
+ }
21
+ }
22
+ ```
23
+ A high number of inserted lines is not necessarily sign of a bad pair, as many translations concatentate or splice source chapters when publishing.
24
+ Instead, watch out for high numbers of missed lines or entries where the inserted liens is high for both source and target.