handler.py

from typing import  Dict, List, Any
# from optimum.onnxruntime import ORTModelForSequenceClassification
# from transformers import pipeline, AutoTokenizer
from FlagEmbedding import BGEM3FlagModel
import time

class EndpointHandler():
    def __init__(self, path="."):
        # load the optimized model
        # モデルの準備
        self.model = BGEM3FlagModel(path, use_fp16=True)


    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 object returned should be a list of one list like [[{"label": 0.9939950108528137}]] containing :
                - "label": A string representing what the label/class is. There can be multiple labels.
                - "score": A score between 0 and 1 describing how confident the model is for this label/class.
        """
        inputs = data.pop("inputs", data)
        parameters = data.pop("parameters", None)
        
        # encodeメソッドの実行前に時間を記録
        start_time = time.time()
        
        result = self.model.encode(inputs, return_dense=False, return_sparse=True, max_length=1024)
        
        # encodeメソッドの実行後に時間を記録
        end_time = time.time()
        # print(result)
        # dense_vectors = result["dense_vecs"]
        
        # 経過時間を計算
        elapsed_time = end_time - start_time
        print(f"Encoding took {elapsed_time:.4f} seconds")
        
        sparse_vectors = result["lexical_weights"]
        # defaultdict(<class 'int'>, {'6': 0.09546, '192661': 0.3323})

        # pass inputs with all kwargs in data
        # if parameters is not None:
        #     prediction = self.pipeline(inputs, **parameters)
        # else:
        #     prediction = self.pipeline(inputs)
        # postprocess the prediction
        
        # レスポンスをの型をkey=str, value=floatのdictにする。なお、numpy.float16はjsonに変換できないので、floatに変換する。
        sparse_vectors = {str(k): float(v) for k, v in sparse_vectors.items()}
        
        # レスポンスの型をnumpy.ndarrayから、通常のarrayに変更する
        # dense_vectors = dense_vectors.tolist()
        
        return [
            [
                { "outputs": sparse_vectors}
            ]
        ]