File size: 1,362 Bytes
c17e24b de4f232 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 |
BART Large CNN model trained for converting NLP queries to SQL queries.
The model was trained on the Spider dataset Link: https://yale-lily.github.io/spider
The model was trained using Google colab.
Hyperparameters:
"num epochs" = 3
"learning rate" = 1e-5
"batch size" = 8
"weight decay" = 0.01
"max input length" = 256
"max target length" = 256
"model name" : "facebook/bart-large-cnn"
use code :
from typing import List
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("upadhyay/sql")
model = AutoModelForSeq2SeqLM.from_pretrained("upadhyay/sql")
def prepare_input(question: str, table: List[str]):
table_prefix = "table:"
question_prefix = "question:"
join_table = ",".join(table)
inputs = f"{question_prefix} {question} {table_prefix} {join_table}"
input_ids = tokenizer(inputs, max_length=700, return_tensors="pt").input_ids
return input_ids
def inference(question: str, table: List[str]) -> str:
input_data = prepare_input(question=question, table=table)
input_data = input_data.to(model.device)
outputs = model.generate(inputs=input_data, num_beams=10, top_k=10, max_length=700)
result = tokenizer.decode(token_ids=outputs[0], skip_special_tokens=True)
return result
print(inference(question="what is salary?", table=["id", "name", "age"]))
|