| from typing import Dict, List, Any | |
| from transformers import AutoTokenizer, AutoModelForSeq2SeqLM | |
| import torch | |
| from handler import EndpointHandler | |
| class EndpointHandler(): | |
| def __init__(self, path=""): | |
| self.tokenizer = AutoTokenizer.from_pretrained(path) | |
| self.model = AutoModelForSeq2SeqLM.from_pretrained(path) | |
| def __call__(self, data: str) -> str: | |
| inp = self.tokenizer(data, return_tensors="pt") | |
| with torch.inference_mode(): | |
| out= self.model.generate(**inp) | |
| final_output = self.tokenizer.batch_decode(out,skip_special_tokens=True) | |
| return {"translation": final_output[0]} |