akeyhero commited on
Commit
0f51d9e
1 Parent(s): 820f89e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +37 -0
README.md CHANGED
@@ -12,6 +12,8 @@ size_categories:
12
  # Overview
13
  This dataset is of conversations extracted from [Aozora Bunko (青空文庫)](https://www.aozora.gr.jp/), which collects public-domain books in Japan, using a simple heuristic approach.
14
 
 
 
15
  # Method
16
  First, lines surrounded by quotation mark pairs (`「」`) are extracted as utterances from the `text` field of [globis-university/aozorabunko-clean](https://huggingface.co/datasets/globis-university/aozorabunko-clean).
17
  Then, consecutive utterances are collected and grouped together.
@@ -24,5 +26,40 @@ As the conversations are extracted using a simple heuristic, a certain amount of
24
  # Tips
25
  If you prefer to employ only modern Japanese, you can filter entries with: `row["meta"]["文字遣い種別"] == "新字新仮名"`.
26
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
27
  # License
28
  CC BY 4.0
 
12
  # Overview
13
  This dataset is of conversations extracted from [Aozora Bunko (青空文庫)](https://www.aozora.gr.jp/), which collects public-domain books in Japan, using a simple heuristic approach.
14
 
15
+ [For Japanese] 日本語での概要説明を Qiita に記載しました: https://qiita.com/akeyhero/items/b53eae1c0bc4d54e321f
16
+
17
  # Method
18
  First, lines surrounded by quotation mark pairs (`「」`) are extracted as utterances from the `text` field of [globis-university/aozorabunko-clean](https://huggingface.co/datasets/globis-university/aozorabunko-clean).
19
  Then, consecutive utterances are collected and grouped together.
 
26
  # Tips
27
  If you prefer to employ only modern Japanese, you can filter entries with: `row["meta"]["文字遣い種別"] == "新字新仮名"`.
28
 
29
+ # Example
30
+
31
+ ```py
32
+ >>> from datasets import load_dataset
33
+ >>> ds = load_dataset('globis-university/aozorabunko-chats')
34
+ >>> ds
35
+ DatasetDict({
36
+ train: Dataset({
37
+ features: ['chats', 'footnote', 'meta'],
38
+ num_rows: 5531
39
+ })
40
+ })
41
+ >>> ds = ds.filter(lambda row: row['meta']['文字遣い種別'] == '新字新仮名') # only modern Japanese
42
+ >>> ds
43
+ DatasetDict({
44
+ train: Dataset({
45
+ features: ['chats', 'footnote', 'meta'],
46
+ num_rows: 4139
47
+ })
48
+ })
49
+ >>> book = ds['train'][0] # one of the works
50
+ >>> book['meta']['作品名']
51
+ 'スリーピー・ホローの伝説'
52
+ >>> chats = book['chats'] # list of the chats in the work; type: list[list[str]]
53
+ >>> len(chats)
54
+ 1
55
+ >>> chat = chats[0] # one of the chats; type: list[str]
56
+ >>> for utterance in chat:
57
+ ... print(utterance)
58
+ ...
59
+ 人生においては、たとえどんな場合でも必ず利点や愉快なことがあるはずです。もっともそれは、わたくしどもが冗談をすなおに受けとればのことですが
60
+ そこで、悪魔の騎士と競走することになった人は、とかくめちゃくちゃに走るのも当然です
61
+ したがって、田舎の学校の先生がオランダ人の世継ぎ娘に結婚を拒まれるということは、彼にとっては、世の中で栄進出世にいたるたしかな一歩だということになります
62
+ ```
63
+
64
  # License
65
  CC BY 4.0