abhishek HF staff commited on
Commit
f134199
1 Parent(s): ca10f72

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +40 -1
README.md CHANGED
@@ -14,4 +14,43 @@ co2_eq_emissions: 39.76330395590446
14
  # Model Trained Using AutoNLP
15
 
16
  - Problem type: Extractive Question Answering
17
- - CO2 Emissions (in grams): 39.76330395590446
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
14
  # Model Trained Using AutoNLP
15
 
16
  - Problem type: Extractive Question Answering
17
+ - CO2 Emissions (in grams): 39.76330395590446
18
+
19
+
20
+ ## Usage
21
+
22
+ You can use cURL to access this model:
23
+
24
+ ```
25
+ $ curl -X POST -H "Authorization: Bearer YOUR_API_KEY" -H "Content-Type: application/json" -d '{"question": "Who loves AutoNLP?", "context": "Everyone loves AutoNLP"}' https://api-inference.huggingface.co/models/abhishek/autonlp-hindi-question-answering-23865268
26
+ ```
27
+
28
+ Or Python API:
29
+
30
+ ```
31
+ import torch
32
+
33
+ from transformers import AutoModelForQuestionAnswering, AutoTokenizer
34
+
35
+ model = AutoModelForQuestionAnswering.from_pretrained("abhishek/autonlp-hindi-question-answering-23865268", use_auth_token=True)
36
+
37
+ tokenizer = AutoTokenizer.from_pretrained("abhishek/autonlp-hindi-question-answering-23865268", use_auth_token=True)
38
+
39
+ from transformers import BertTokenizer, BertForQuestionAnswering
40
+
41
+ question, text = "Who loves AutoNLP?", "Everyone loves AutoNLP"
42
+
43
+ inputs = tokenizer(question, text, return_tensors='pt')
44
+
45
+ start_positions = torch.tensor([1])
46
+
47
+ end_positions = torch.tensor([3])
48
+
49
+ outputs = model(**inputs, start_positions=start_positions, end_positions=end_positions)
50
+
51
+ loss = outputs.loss
52
+
53
+ start_scores = outputs.start_logits
54
+
55
+ end_scores = outputs.end_logits
56
+ ```