# model_api.py

from fastapi import FastAPI
from pydantic import BaseModel

app = FastAPI()

# Define the input data model for your AI API
class EquipmentInput(BaseModel):
    usage_hours: float
    idle_hours: float
    movement_frequency: float
    cost_per_hour: float
    last_maintenance: str  # Expecting 'YYYY-MM-DD' format

# Example GET endpoint (test your server)
@app.get("/")
async def root():
    return {"message": "AI model API is up and running!"}

# Example POST endpoint to accept input and respond with dummy AI results
@app.post("/predict")
async def predict(input_data: EquipmentInput):
    # Here you would call your real AI model logic
    # For now, returning dummy data
    suggestion = "Move"
    confidence = 0.92
    utilization_score = 0.88
    return {
        "suggestion": suggestion,
        "confidence": confidence,
        "utilization_score": utilization_score,
        "input_received": input_data.dict()
    }