Mizuiro-sakura commited on
Commit
3fec28b
·
1 Parent(s): d181b55

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +10 -2
README.md CHANGED
@@ -20,16 +20,24 @@ This could be able to distinguish between positive and negative content.
20
  This model was fine-tuned by using Natsume Souseki's documents.
21
  For example Kokoro, Bocchan, Sanshiro and so on...
22
 
23
- # what is Luke?[1]
24
  LUKE (Language Understanding with Knowledge-based Embeddings) is a new pre-trained contextualized representation of words and entities based on transformer. LUKE treats words and entities in a given text as independent tokens, and outputs contextualized representations of them. LUKE adopts an entity-aware self-attention mechanism that is an extension of the self-attention mechanism of the transformer, and considers the types of tokens (words or entities) when computing attention scores.
25
 
26
  LUKE achieves state-of-the-art results on five popular NLP benchmarks including SQuAD v1.1 (extractive question answering), CoNLL-2003 (named entity recognition), ReCoRD (cloze-style question answering), TACRED (relation classification), and Open Entity (entity typing).
 
27
 
 
28
  # how to use 使い方
29
  出力としてはpre.logitsが得られます。
30
- pre.logitsはtensor[[x,y]]というテンソルになっています。
31
  num = SOFTMAX(pre.logits)にすることで、num[0]がネガティブである確率、num[1]がポジティブである確率を表すようになります。
32
 
 
 
 
 
 
 
33
  -------------------------------------------------------------
34
 
35
  import torch
 
20
  This model was fine-tuned by using Natsume Souseki's documents.
21
  For example Kokoro, Bocchan, Sanshiro and so on...
22
 
23
+ # what is Luke? Lukeとは?[1]
24
  LUKE (Language Understanding with Knowledge-based Embeddings) is a new pre-trained contextualized representation of words and entities based on transformer. LUKE treats words and entities in a given text as independent tokens, and outputs contextualized representations of them. LUKE adopts an entity-aware self-attention mechanism that is an extension of the self-attention mechanism of the transformer, and considers the types of tokens (words or entities) when computing attention scores.
25
 
26
  LUKE achieves state-of-the-art results on five popular NLP benchmarks including SQuAD v1.1 (extractive question answering), CoNLL-2003 (named entity recognition), ReCoRD (cloze-style question answering), TACRED (relation classification), and Open Entity (entity typing).
27
+ luke-japaneseは、単語とエンティティの知識拡張型訓練済み Transformer モデルLUKEの日本語版です。LUKE は単語とエンティティを独立したトークンとして扱い、これらの文脈を考慮した表現を出力します。詳細については、GitHub リポジトリを参照してください。
28
 
29
+ このモデルは、通常の NLP タスクでは使われない Wikipedia エンティティのエンベディングを含んでいます。単語の入力のみを使うタスクには、lite versionを使用してください。
30
  # how to use 使い方
31
  出力としてはpre.logitsが得られます。
32
+ pre.logitsはtensor[[x, y]]というテンソルになっています。
33
  num = SOFTMAX(pre.logits)にすることで、num[0]がネガティブである確率、num[1]がポジティブである確率を表すようになります。
34
 
35
+ we could get "pre.logits" as the output.
36
+ "pre.logits" is the shape like tensor[[x, y]].
37
+ "num = SOFTMAX(pre.logits)"
38
+ num[0] will show the probability of negative, num[1] will show the probability of positive.
39
+
40
+
41
  -------------------------------------------------------------
42
 
43
  import torch