handler.py

from typing import Dict, List, Any
from transformers import AutoTokenizer, AutoModel
import torch

#Mean Pooling - Take attention mask into account for correct averaging
def mean_pooling(model_output, attention_mask):
    token_embeddings = model_output[0] #First element of model_output contains all token embeddings
    input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
    return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
    
class EndpointHandler():
    def __init__(self, path=""):
        self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
        self.tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/msmarco-MiniLM-L-6-v3')
        self.model = AutoModel.from_pretrained('sentence-transformers/msmarco-MiniLM-L-6-v3')
        self.model.to(self.device)
        print("model will run on ", self.device)

    def __call__(self, data: Dict[str, Any]) -> List[Dict[str, Any]]:
        """
       data args:
            inputs (:obj: `str` | `PIL.Image` | `np.array`)
            kwargs
      Return:
            A :obj:`list` | `dict`: will be serialized and returned
        """
        sentences = data.pop("inputs",data)
        encoded_input = self.tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
        encoded_input = {key: value.to(self.device) for key, value in encoded_input.items()}
        # Compute token embeddings
        with torch.no_grad():
            model_output = self.model(**encoded_input)

        # Perform pooling. In this case, max pooling.
        sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
        return sentence_embeddings.tolist()