app.py

from fastapi import FastAPI, HTTPException
from pydantic import BaseModel
import joblib
import pandas as pd
import logging

# Set up logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

# Load the model
try:
    model = joblib.load("titanic_model.pkl")
    logger.info(f"Model loaded successfully. Feature names: {model.feature_names_in_}")
except Exception as e:
    logger.error(f"Error loading model: {e}")
    raise

# Create the Pydantic model for the input data
class Passenger(BaseModel):
    pclass: int
    sex: str
    age: float
    sibsp: int
    parch: int
    fare: float
    embarked: str


# {
#   "pclass": 1,
#   "sex": "male",
#   "age": 30,
#   "sibsp": 0,
#   "parch": 0,
#   "fare": 100,
#   "embarked": "S"
# }

# Create the FastAPI instance
app = FastAPI()

# Create the root endpoint
@app.get("/")
def read_root():
    return {"message": "Welcome to the Titanic Survival Prediction API"}

# Create the predict endpoint
@app.post("/predict")
def predict(passenger: Passenger):
    try:
        # Convert the input data to a DataFrame
        input_dict = passenger.model_dump()
        logger.info(f"Input data: {input_dict}")
        
        input_data = pd.DataFrame([input_dict])
        logger.info(f"DataFrame created with columns: {input_data.columns.tolist()}")
        
        # One-Hot Encode the input data
        input_data = pd.get_dummies(input_data)
        logger.info(f"After one-hot encoding, columns: {input_data.columns.tolist()}")
        
        # Check if model has feature_names_in_ attribute
        if not hasattr(model, 'feature_names_in_'):
            raise HTTPException(status_code=500, detail="Model does not have feature_names_in_ attribute")
        
        logger.info(f"Model expects columns: {model.feature_names_in_}")
        
        # Align the input data columns with the model columns
        input_data = input_data.reindex(columns=model.feature_names_in_, fill_value=0)
        logger.info(f"After reindexing, columns: {input_data.columns.tolist()}")
        
        # Check if we have the right number of features
        if input_data.shape[1] != len(model.feature_names_in_):
            raise HTTPException(
                status_code=500, 
                detail=f"Feature mismatch: Input has {input_data.shape[1]} features, model expects {len(model.feature_names_in_)}"
            )
        
        # Predict the survival of the passenger
        prediction = model.predict(input_data)
        
        return {
            "prediction": int(prediction[0]),
            "prediction_probability": float(model.predict_proba(input_data)[0][1]) if hasattr(model, 'predict_proba') else None
        }
        
    except Exception as e:
        logger.error(f"Prediction error: {e}")
        raise HTTPException(status_code=500, detail=f"Prediction failed: {str(e)}")