handler.py

from typing import  Dict, List, Any

import torch
import torch.nn.functional as F
from transformers import AutoTokenizer
from optimum.onnxruntime import ORTModelForFeatureExtraction


def mean_pooling(model_output, attention_mask):
    token_embeddings = model_output[0]
    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.tokenizer = AutoTokenizer.from_pretrained(path)
        self.model = ORTModelForFeatureExtraction.from_pretrained(path, file_name="model.onnx")
        
    def __call__(self, data: Any) -> List[List[Dict[str, float]]]:
        """
        Args:
            data (:obj:):
                includes the input data and the parameters for the inference.
        Return:
            A :obj:`list`:. The list contains the embeddings of the inference inputs
        """
        inputs = data.get("inputs", data)

        # tokenize the input
        encoded_inputs = self.tokenizer(inputs, padding=True, truncation=True, return_tensors='pt')
        # run the model
        outputs = self.model(**encoded_inputs)
        # Perform pooling
        sentence_embeddings = mean_pooling(outputs, encoded_inputs['attention_mask'])
        # Normalize embeddings
        sentence_embeddings = F.normalize(sentence_embeddings, p=2, dim=1)
        # postprocess the prediction
        return {"embeddings": sentence_embeddings.tolist()}