src/streamlit_app.py

import streamlit as st
import joblib
import pandas as pd
import numpy as np

MODEL_PATH = 'src/vgsales_prediction.joblib' 
FEATURES = ["Rank", "NA_Sales", "EU_Sales", "JP_Sales", "Other_Sales"]

@st.cache_resource
def load_linear_model():
    try:
        model = joblib.load(MODEL_PATH)
        return model
    except Exception as e:
        st.error(f"Error loading the Linear Regression model. Ensure '{MODEL_PATH}' is uploaded. Error: {e}")
        return None

def predict_sales(model, input_data):
    input_df = pd.DataFrame([input_data])[FEATURES]
    
    prediction = model.predict(input_df)
    
    return float(prediction[0])

# --- Streamlit Interface ---
st.set_page_config(page_title="Game Sales Predictor", layout="centered")
st.title("🎮 Global Video Game Sales Prediction")
st.markdown("Enter regional sales figures (in Millions) and Rank to predict total Global Sales.")

model = load_linear_model()

if model is not None:
    st.sidebar.header("Sales Input (in Millions)")

    rank = st.sidebar.number_input("Game Rank:", min_value=1, value=1000)
    
    na_sales = st.sidebar.number_input("North America Sales (NA_Sales):", min_value=0.0, value=0.5, step=0.01)
    eu_sales = st.sidebar.number_input("Europe Sales (EU_Sales):", min_value=0.0, value=0.25, step=0.01)
    jp_sales = st.sidebar.number_input("Japan Sales (JP_Sales):", min_value=0.0, value=0.1, step=0.01)
    other_sales = st.sidebar.number_input("Other Sales (Other_Sales):", min_value=0.0, value=0.05, step=0.01)

    input_data = {
        "Rank": rank,
        "NA_Sales": na_sales,
        "EU_Sales": eu_sales,
        "JP_Sales": jp_sales,
        "Other_Sales": other_sales
    }

    st.subheader("Regional Sales Input (Millions):")
    st.dataframe(pd.DataFrame([input_data]), hide_index=True)

    if st.button("Predict Global Sales"):
        with st.spinner('Calculating prediction...'):
            predicted_global_sales = predict_sales(model, input_data)
            
            st.success("Prediction Successful!")
            st.markdown("### Predicted Global Sales:")
            st.markdown(f"**{predicted_global_sales:,.2f} Million**")
            
            st.info("Note: This simple linear model uses Rank and regional sales to predict the total.")