prediction.py

import streamlit as st
import pandas as pd
import pickle  # or import pickle if that's what you used to save your model

# Load your trained model
with open('XGBmodel.pkl', 'rb') as file_1:
    XGBmodel = pickle.load(file_1)

# Set the title of the app
def  make_prediction():
    st.title('Laptop Price Prediction')
    st.write("MLModel Built by MSABP")
    st.header('',divider='gray')

    # Create a form for inputting laptop specifications
    with st.form(key='laptop_spec_form'):
        # Input fields
        company = st.text_input("Company")
        product = st.text_input("Product")
        laptop_type = st.selectbox("Type", ["Notebook", "Gaming", "Ultrabook", "2-in-1", "Others"])
        inches = st.number_input("Inches", min_value=0, step=1)
        ram = st.number_input("RAM (GB)", min_value=0, step=2)
        os = st.text_input("Operating System")
        weight = st.number_input("Weight (kg)", min_value=0.0, step=0.1)
        screen = st.selectbox("Screen Resolution", ["Full HD", "4K", "HD", "Others"])
        width = st.number_input("Screen Width (px)", min_value=0, step=120)
        height = st.number_input("Screen Height (px)", min_value=0, step=120)
        touchscreen = st.selectbox("Touchscreen", ["Yes", "No"])
        ips = st.selectbox("IPS Display", ["Yes", "No"])
        retina = st.selectbox("Retina Display", ["Yes", "No"])
        cpu_company = st.text_input("CPU Company")
        cpu_freq = st.number_input("CPU Frequency (GHz)", min_value=0.0, step=0.1)
        cpu_model = st.text_input("CPU Model")
        primary_storage = st.number_input("Primary Storage (GB)", min_value=0, step=64)
        secondary_storage = st.number_input("Secondary Storage (GB)", min_value=0, step=64)
        primary_storage_type = st.selectbox("Primary Storage Type", ["SSD", "HDD", "Others"])
        secondary_storage_type = st.selectbox("Secondary Storage Type", ["No", "SSD", "HDD", "Others"])
        gpu_company = st.text_input("GPU Company")
        gpu_model = st.text_input("GPU Model")

        # Submit button
        submit_button = st.form_submit_button(label='Submit')

        if submit_button:
            # Create a dictionary from the input data
            laptop_data = {
                "Company": company,
                "Product": product,
                "Type": laptop_type,
                "Inches": inches,
                "Ram": ram,
                "OS": os,
                "Weight": weight,
                "Screen": screen,
                "Width": width,
                "Height": height,
                "Touchscreen": touchscreen,
                "IPS": ips,
                "Retina": retina,
                "CPU_company": cpu_company,
                "CPU_freq": cpu_freq,
                "CPU_model": cpu_model,
                "PrimaryStorage": primary_storage,
                "SecondaryStorage": secondary_storage,
                "PrimaryStorageType": primary_storage_type,
                "SecondaryStorageType": secondary_storage_type,
                "GPU_company": gpu_company,
                "GPU_model": gpu_model
            }

            # Convert the input data to a DataFrame for prediction
            input_df = pd.DataFrame([laptop_data])

            st.subheader('User Input')
            st.write("""This is a view of the data you have entered.
            """)
            st.write(input_df)

            # Make the prediction
            predicted_price = XGBmodel.predict(input_df)

            # Display the prediction message
            st.write(f'Based on specs you wanted, the price estimation is: {predicted_price[0]:,.2f}')



if __name__ == "__main__":
    make_prediction()