| from typing import Any, Dict, List | |
| import torch | |
| import transformers | |
| from transformers import AutoModelForCausalLM, AutoTokenizer | |
| dtype = torch.bfloat16 if torch.cuda.get_device_capability()[0] ==8 else torch.float16 | |
| class EndpointHandler: | |
| def __init__(self, path=""): | |
| self.tokenizer = AutoTokenizer.from_pretrained(path) | |
| self.model = AutoModelForCausalLM.from_pretrained(path, trust_remote_code=True, revision="main") | |
| self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu") | |
| self.model = self.model.to(self.device) | |
| def __call__(self, data: Dict[str, Any]) -> Dict[str, Any]: | |
| prompt = data["inputs"] | |
| if "config" in data: | |
| config = data.pop("config", None) | |
| else: | |
| config = {'max_new_tokens':100} | |
| input_ids = self.tokenizer(prompt, return_tensors="pt").input_ids.to(self.device) | |
| generated_ids = self.model.generate(input_ids, **config) | |
| generated_text = self.tokenizer.decode(generated_ids[0], skip_special_tokens=True) | |
| return [{"generated_text": generated_text}] | |