hantian2 commited on
Commit
82c804e
1 Parent(s): 44bc19d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +20 -0
README.md CHANGED
@@ -2,11 +2,31 @@
2
  license: apache-2.0
3
  ---
4
 
 
 
 
 
 
5
  # Usage
6
 
 
 
7
  ```
8
  from transformers import AutoTokenizer, AutoModelForSequenceClassification
 
9
 
10
  tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
11
  model = AutoModelForSequenceClassification.from_pretrained("CogComp/ZeroShotWiki")
 
 
 
 
 
 
 
 
 
 
 
 
12
  ```
 
2
  license: apache-2.0
3
  ---
4
 
5
+ # Model description
6
+
7
+ A BertForSequenceClassification model that is finetuned on Wikipedia for zero-shot text classification. For details, see our NAACL'22 paper.
8
+
9
+
10
  # Usage
11
 
12
+ Concatenate the text sentence with each of the candidate labels as input to the model. The model will output a score for each label. Below is an example.
13
+
14
  ```
15
  from transformers import AutoTokenizer, AutoModelForSequenceClassification
16
+ import torch
17
 
18
  tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
19
  model = AutoModelForSequenceClassification.from_pretrained("CogComp/ZeroShotWiki")
20
+
21
+ labels = ["sports", "business", "politics"]
22
+ texts = ["As of the 2018 FIFA World Cup, twenty-one final tournaments have been held and a total of 79 national teams have competed."]
23
+
24
+ with torch.no_grad():
25
+ for text in texts:
26
+ label_score = {}
27
+ for label in labels:
28
+ inputs = tokenizer(text, label, return_tensors='pt')
29
+ out = model(**inputs)
30
+ label_score[label]=float(torch.nn.functional.softmax(out[0], dim=-1)[0][0])
31
+ print(label_score) # Predict the label with the highest score
32
  ```