app.py

from transformers import AutoTokenizer
from onnxruntime import InferenceSession
import numpy as np
import json
from fastapi import FastAPI

app = FastAPI()

# Initialize components
tokenizer = AutoTokenizer.from_pretrained(
    "Xenova/multi-qa-mpnet-base-dot-v1",
    use_fast=False  # Avoids framework dependencies
)
session = InferenceSession("model.onnx")

def cosine_similarity(a, b):
    return np.dot(a, b) / (np.linalg.norm(a) * np.linalg.norm(b))

@app.post("/predict")
async def predict(query: str):
    # Tokenize
    inputs = tokenizer(query, return_tensors="np")
    inputs = {k: v.astype(np.int64) for k, v in inputs.items()}
    
    # Get embedding
    embedding = session.run(None, inputs)[0][0]
    
    return {"embedding": embedding.tolist()}