Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
|
3 |
+
license: apache-2.0
|
4 |
+
inference: false
|
5 |
+
|
6 |
+
---
|
7 |
+
|
8 |
+
|
9 |
+
```
|
10 |
+
from transformers import AutoTokenizer, AutoModelWithLMHead, AutoModelForCausalLM
|
11 |
+
import torch
|
12 |
+
if torch.cuda.is_available():
|
13 |
+
device = torch.device("cuda")
|
14 |
+
else :
|
15 |
+
device = "cpu"
|
16 |
+
|
17 |
+
|
18 |
+
tokenizer = AutoTokenizer.from_pretrained("salesken/grammar_correction")
|
19 |
+
model = AutoModelForCausalLM.from_pretrained("salesken/grammar_correction").to(device)
|
20 |
+
|
21 |
+
input_query="what be the reason for everyone leave the company"
|
22 |
+
query= "<|startoftext|> " + input_query + " ~~~"
|
23 |
+
|
24 |
+
|
25 |
+
input_ids = tokenizer.encode(query.lower(), return_tensors='pt').to(device)
|
26 |
+
sample_outputs = model.generate(input_ids,
|
27 |
+
do_sample=True,
|
28 |
+
num_beams=1,
|
29 |
+
max_length=128,
|
30 |
+
temperature=0.9,
|
31 |
+
top_p= 0.7,
|
32 |
+
top_k = 5,
|
33 |
+
num_return_sequences=3)
|
34 |
+
corrected_sentences = []
|
35 |
+
for i in range(len(sample_outputs)):
|
36 |
+
r = tokenizer.decode(sample_outputs[i], skip_special_tokens=True).split('||')[0]
|
37 |
+
r = r.split('~~~')[1]
|
38 |
+
if r not in corrected_sentences:
|
39 |
+
corrected_sentences.append(r)
|
40 |
+
|
41 |
+
print(corrected_sentences)
|
42 |
+
```
|