ThomasNLG commited on
Commit
0309c11
1 Parent(s): 09efe72

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -6
README.md CHANGED
@@ -2,7 +2,7 @@
2
  language: en
3
  tags:
4
  - qa
5
- - question
6
  - answering
7
  - SQuAD
8
  - data2text
@@ -15,16 +15,16 @@ datasets:
15
  model-index:
16
  - name: t5-qa_webnlg_synth-en
17
  results:
18
- - task:
19
  name: Data Question Answering
20
  type: extractive-qa
21
  widget:
22
- - text: "What is the food type at The Eagle? </s> name [ The Eagle ] , eatType [ coffee shop ] , food [ French ] , priceRange [ £ 2 0 - 2 5 ]"
23
  ---
24
  # t5-qa_webnlg_synth-en
25
 
26
  ## Model description
27
- This model is a *Data Question Answering* model based on T5-small, that answer questions given a structured table as input.
28
  It is actually a component of [QuestEval](https://github.com/recitalAI/QuestEval) metric but can be used independently as it is, for QA only.
29
 
30
 
@@ -43,7 +43,7 @@ You can play with the model using the inference API, the text input format shoul
43
 
44
  where CONTEXT is a structured table that is linearised this way:
45
 
46
- `CONTEXT = "name [ The Eagle ] , eatType [ coffee shop ] , food [ French ] , priceRange [ £ 2 0 - 2 5 ]"`
47
 
48
 
49
  ## Training data
@@ -58,5 +58,4 @@ The model was trained on synthetic data as described in [Data-QuestEval: A Refer
58
  journal={arXiv preprint arXiv:2104.07555},
59
  year={2021}
60
  }
61
- }
62
  ```
 
2
  language: en
3
  tags:
4
  - qa
5
+ - question
6
  - answering
7
  - SQuAD
8
  - data2text
 
15
  model-index:
16
  - name: t5-qa_webnlg_synth-en
17
  results:
18
+ - task:
19
  name: Data Question Answering
20
  type: extractive-qa
21
  widget:
22
+ - text: "What is the food type at The Eagle? </s> name [ The Eagle ] , eatType [ coffee shop ] , food [ French ] , priceRange [ £ 2 0 - 2 5 ]"
23
  ---
24
  # t5-qa_webnlg_synth-en
25
 
26
  ## Model description
27
+ This model is a *Data Question Answering* model based on T5-small, that answers questions given a structured table as input.
28
  It is actually a component of [QuestEval](https://github.com/recitalAI/QuestEval) metric but can be used independently as it is, for QA only.
29
 
30
 
 
43
 
44
  where CONTEXT is a structured table that is linearised this way:
45
 
46
+ `CONTEXT = "name [ The Eagle ] , eatType [ coffee shop ] , food [ French ] , priceRange [ £ 2 0 - 2 5 ]"`
47
 
48
 
49
  ## Training data
 
58
  journal={arXiv preprint arXiv:2104.07555},
59
  year={2021}
60
  }
 
61
  ```