Gestructureerde voorspellingsrespons retourneren

AI in productie brengen met FastAPI

Matt Eckerle

Software and Data Engineering Leader

Uitdagingen bij het uitrollen van modellen

Stroomschema met ML-voorspelling in FastAPI

Accepteer invoergegevens correct
Valideer inkomende data en handel fouten af
Maak voorspellingen
Retourneer goed gestructureerde responses

Verzoekstructuur definiëren

from pydantic import BaseModel

class PredictionRequest(BaseModel):
    text: str

class PredictionResponse(BaseModel):

    text: str

    sentiment: str

    confidence: float

De prediction-endpoint maken

@app.post("/predict")
def predict_sentiment(request: PredictionRequest):
    if sentiment_model is None:
        raise HTTPException(
            status_code=503,
            detail="Model not loaded"
        )

    result = sentiment_model(request.text)

    return PredictionResponse(
            text=request.text,
            sentiment=result[0]["label"],
            confidence=result[0]["score"]
        )

Input-JSON:

{"text": "This movie was fantastic!"}

Response:

{
    "text": "This movie was fantastic!",
    "sentiment": "POSITIVE",
    "confidence": 0.95
}

Fouthandling

try:
  result = sentiment_model(request.text)
  return PredictionResponse(
        text=request.text,
        sentiment=result[0]["label"],
        confidence=result[0]["score"]
    )
except Exception:
  raise HTTPException(
        status_code=500,
        detail="Prediction failed"
    )

Response wanneer het model geen voorspelling kan doen

{
    "detail": "Prediction failed",
    "status_code": 500
}

De endpoint testen

# Example request
import requests

response = requests.post(
    "http://localhost:8000/predict",
    json={"text": "Great product!"}
)
print(response.json())

{
    "text": "Great product!",
    "sentiment": "POSITIVE",
    "confidence": 0.998
}

Laten we oefenen!

AI in productie brengen met FastAPI