StevenTang commited on
Commit
c2a8242
1 Parent(s): 52baadf

Update README

Browse files
Files changed (1) hide show
  1. README.md +42 -0
README.md ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ language:
4
+ - en
5
+ tags:
6
+ - text-generation
7
+ - text2text-generation
8
+ pipeline_tag: text2text-generation
9
+ widget:
10
+ - text: "Describe the following data: Iron Man | instance of | Superhero [SEP] Stan Lee | creator | Iron Man"
11
+ example_title: "Example1"
12
+ - text: "Describe the following data: First Clearing | LOCATION | On NYS 52 1 Mi. Youngsville [SEP] On NYS 52 1 Mi. Youngsville | CITY_OR_TOWN | Callicoon, New York"
13
+ example_title: "Example2"
14
+ ---
15
+
16
+ # MTL-data-to-text
17
+ The MTL-data-to-text model was proposed in [**MVP: Multi-task Supervised Pre-training for Natural Language Generation**](https://github.com/RUCAIBox/MVP/blob/main/paper.pdf) by Tianyi Tang, Junyi Li, Wayne Xin Zhao and Ji-Rong Wen.
18
+
19
+ The detailed information and instructions can be found [https://github.com/RUCAIBox/MVP](https://github.com/RUCAIBox/MVP).
20
+
21
+ ## Model Description
22
+ MTL-data-to-text is supervised pre-trained using a mixture of labeled data-to-text datasets. It is a variant (Single) of our main MVP model. It follows a standard Transformer encoder-decoder architecture.
23
+
24
+ MTL-data-to-text is specially designed for data-to-text generation tasks, such as KG-to-text generation (WebNLG, DART), table-to-text generation (WikiBio, ToTTo) and MR-to-text generation (E2E).
25
+
26
+ ## Example
27
+ ```python
28
+ >>> from transformers import MvpTokenizer, MvpForConditionalGeneration
29
+
30
+ >>> tokenizer = MvpTokenizer.from_pretrained("RUCAIBox/mvp")
31
+ >>> model = MvpForConditionalGeneration.from_pretrained("RUCAIBox/mtl-data-to-text")
32
+
33
+ >>> inputs = tokenizer(
34
+ ... "Describe the following data: Iron Man | instance of | Superhero [SEP] Stan Lee | creator | Iron Man",
35
+ ... return_tensors="pt",
36
+ ... )
37
+ >>> generated_ids = model.generate(**inputs)
38
+ >>> tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
39
+ ['Iron Man is a fictional superhero appearing in American comic books published by Marvel Comics.']
40
+ ```
41
+
42
+ ## Citation