api.py

# app.py
from fastapi import FastAPI
from pydantic import BaseModel
from typing import Literal
import numpy as np
import joblib


# Load all components
model = joblib.load("model_GBR_custom.pkl")
scaler = joblib.load("scaler.pkl")
encoders = joblib.load("encoders.pkl")

app = FastAPI()

# Define input schema
class CropInput(BaseModel):
    crop: str
    season: str
    ndvi: float
    lst: float
    windspeed: float
    temperature: float
    total_rainfall: float
    relative_humidity: float
    rsm: float
    N: float
    P: float
    K: float

@app.post("/predict")
def predict(data: CropInput):
    # Extract fields
    x = [
        data.crop,
        data.season,
        data.ndvi,
        data.lst,
        data.windspeed,
        data.temperature,
        data.total_rainfall,
        data.relative_humidity,
        data.rsm,
        data.N,
        data.P,
        data.K
    ]

    # Apply encoders (assuming encoders is a dict with keys 'crop' and 'season')
    x[0] = encoders["crop"].transform([x[0]])[0]
    x[1] = encoders["season"].transform([x[1]])[0]

    # Convert to array and scale
    x_arr = np.array(x).reshape(1, -1)
    x_scaled = scaler.transform(x_arr)

    # Predict
    prediction = model.predict(x_scaled)[0]
    return {"prediction": prediction}

# To run locally:
# uvicorn app:app --reload