Commit
·
73a545c
1
Parent(s):
5c0597d
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
#import transformers
|
2 |
+
from transformers import (
|
3 |
+
T5ForConditionalGeneration,
|
4 |
+
T5Tokenizer,
|
5 |
+
)
|
6 |
+
|
7 |
+
#load model
|
8 |
+
model = T5ForConditionalGeneration.from_pretrained('dsivakumar/text2sql')
|
9 |
+
tokenizer = T5Tokenizer.from_pretrained('dsivakumar/text2sql')
|
10 |
+
|
11 |
+
#predict function
|
12 |
+
def get_sql(query,tokenizer,model):
|
13 |
+
source_text= "English to SQL: "+query
|
14 |
+
source_text = ' '.join(source_text.split())
|
15 |
+
source = tokenizer.batch_encode_plus([source_text],max_length= 128, pad_to_max_length=True, truncation=True, padding="max_length", return_tensors='pt')
|
16 |
+
source_ids = source['input_ids'] #.squeeze()
|
17 |
+
source_mask = source['attention_mask']#.squeeze()
|
18 |
+
generated_ids = model.generate(
|
19 |
+
input_ids = source_ids.to(dtype=torch.long),
|
20 |
+
attention_mask = source_mask.to(dtype=torch.long),
|
21 |
+
max_length=150,
|
22 |
+
num_beams=2,
|
23 |
+
repetition_penalty=2.5,
|
24 |
+
length_penalty=1.0,
|
25 |
+
early_stopping=True
|
26 |
+
)
|
27 |
+
preds = [tokenizer.decode(g, skip_special_tokens=True, clean_up_tokenization_spaces=True) for g in generated_ids]
|
28 |
+
return preds
|
29 |
+
|
30 |
+
#test
|
31 |
+
query="Show me the average age of of wines in Italy by provinces"
|
32 |
+
sql = get_sql(query,tokenizer,model)
|
33 |
+
print(sql)
|