Update app.py

app.py

@@ -1,35 +1,54 @@
 from fastapi import FastAPI, Request
 from prophet import Prophet
 import pandas as pd
+import logging
 
 app = FastAPI()
 
-# Load the historical data CSV file
-df = pd.read_csv("Queue_Historical_Data_Jan_to_May_2025.csv")
+# Setup logging
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
 
-# Ensure the 'ds' column is in datetime format
-df['ds'] = pd.to_datetime(df['ds'], errors='coerce', dayfirst=False)
+@app.on_event("startup")
+async def startup_event():
+    logger.info("Starting the application")
 
-# Drop rows where 'ds' could not be parsed
-df = df.dropna(subset=['ds'])
+@app.on_event("shutdown")
+async def shutdown_event():
+    logger.info("Shutting down the application")
 
-# Ensure the 'y' column exists and is numerical
-df['y'] = pd.to_numeric(df['y'], errors='coerce')
-df = df.dropna(subset=['y'])
-
-# Initialize the Prophet model
-model = Prophet()
+try:
+    # Load the historical data CSV file
+    df = pd.read_csv("Queue_Historical_Data_Jan_to_May_2025.csv")
+    
+    # Ensure the 'ds' column is in datetime format
+    df['ds'] = pd.to_datetime(df['ds'], errors='coerce', dayfirst=False)
+    df = df.dropna(subset=['ds'])
+    
+    # Ensure the 'y' column exists and is numerical
+    df['y'] = pd.to_numeric(df['y'], errors='coerce')
+    df = df.dropna(subset=['y'])
+    
+    # Initialize the Prophet model
+    model = Prophet()
+    model.fit(df)
+    logger.info("Model successfully fitted")
 
-# Fit the model with historical data
-model.fit(df)
+except Exception as e:
+    logger.error(f"Error during initialization: {e}")
 
 @app.post("/run")
 async def forecast(request: Request):
-    # Create a future dataframe for the next 7 days
-    future = model.make_future_dataframe(df, periods=7)
-    
-    # Predict the future values
-    forecast = model.predict(future)
+    try:
+        # Create a future dataframe for the next 7 days
+        future = model.make_future_dataframe(df, periods=7)
+        
+        # Predict the future values
+        forecast = model.predict(future)
+        
+        # Return the last 7 predicted values
+        return forecast[['ds', 'yhat']].tail(7).to_dict(orient="records")
     
-    # Return the last 7 predicted values
-    return forecast[['ds', 'yhat']].tail(7).to_dict(orient="records")
+    except Exception as e:
+        logger.error(f"Error during forecast: {e}")
+        return {"error": "Error occurred while making predictions"}