sentiment_analyzer/api.py

from typing import Dict

from fastapi import Depends, FastAPI
from pydantic import BaseModel

from .classifier.model import Model, get_model

app = FastAPI()


class SentimentRequest(BaseModel):
    text: str

class SentimentResponse(BaseModel):
    probabilities: Dict[str, float]
    sentiment: str
    confidence: float


@app.post("/predict", response_model=SentimentResponse)
def predict(request: SentimentRequest, model: Model = Depends(get_model)):
    sentiment, confidence, probabilities = model.predict(request.text)
    return SentimentResponse(
        sentiment=sentiment, confidence=confidence, probabilities=probabilities
    )


@app.get("/")
def read_root():
    return {"TrueFoundry": "Project"}