# import streamlit as st
# import pandas as pd
# import pickle

# # Load your trained model #pickle.load() requires a file object opened in binary read mode ('rb').
# with open('models/sales_prediction_pipeline.pkl', 'rb') as file:
#     model = pickle.load(file)

# # Function to predict sales
# def predict_sales(input_data):
#     # Make predictions using the loaded model
#     sales_prediction = model.predict(input_data)
#     return sales_prediction

# # ///////////////////////////////////////////// Streamlit app //////////////////////////////////////////

# def main():
#     st.title('Sales Prediction App')
#     st.image("images\\r1.jpg", caption="Rossmann")


#     # Input widgets
#     PromoInterval = st.selectbox("Promo Interval", ['No Promotion', 'Jan,Apr,Jul,Oct', 'Feb,May,Aug,Nov', 'Mar,Jun,Sept,Dec'])

#     # -----------------------------------------------------------------------------------------------
#     StoreType = st.radio("StoreType", ["Small Shop", "Medium Store", "Large Store", "Hypermarket"])
#     Assortment = st.radio("Assortment", ["basic", "extra", "extended"])
    

#     # Encode StateHoliday as 1 for 'Yes' and 0 for 'No' --------------------------------------
#     StateHoliday = st.radio("State Holiday", ["Yes", "No"])
#     StateHoliday = 1 if StateHoliday == "Yes" else 0

#     SchoolHoliday = st.radio("School Holiday", ["Yes", "No"])
#     SchoolHoliday = 1 if SchoolHoliday == "Yes" else 0

#     Promo = st.radio("Promotion", ["store is participating", "store is not participating"])
#     Promo = 1 if Promo == "store is participating" else 0
#     # ----------------------------------------------------------------------------------------
    

#     Store = st.slider("Store", 1, 1115)
#     Customers = st.slider("Customers", 0, 7388)
#     CompetitionDistance = st.slider("Competition Distance", 20, 75860)
#     CompetitionOpenSinceMonth = st.slider("Competition Open Since Month", 1, 12)
#     CompetitionOpenSinceYear = st.slider("Competition Open Since Year", 1998, 2015)
#     # ----------------------------------------------------------------------------------------

#     # Store user inputs
#     input_data = pd.DataFrame({
#         'PromoInterval': [PromoInterval],
#         'StoreType': [StoreType],
#         'Assortment': [Assortment],
#         'StateHoliday': [StateHoliday],
#         'Store': [Store],
#         'Customers': [Customers],
#         'Promo': [Promo],
#         'SchoolHoliday': [SchoolHoliday],
#         'CompetitionDistance': [CompetitionDistance],
#         'CompetitionOpenSinceMonth': [CompetitionOpenSinceMonth],
#         'CompetitionOpenSinceYear': [CompetitionOpenSinceYear]
#     })

#     # Display input data
#     st.subheader('Input Data:')
#     st.write(input_data)

#     # Predict sales
#     if st.button('Predict Sales'):
#         prediction = predict_sales(input_data)[0]
#         formatted_prediction = "{:.2f}".format(prediction)  # Format prediction to display two decimal points
#         st.write('Predicted Sales:', formatted_prediction)


# if __name__ == '__main__':
#     main()



# # Record at index 795018:
# # PromoInterval                Jan,Apr,Jul,Oct
# # StoreType                         Small Shop
# # Assortment                             basic
# # StateHoliday                               0
# # SchoolHoliday                              0
# # Promo                                      1
# # Store                                    650
# # Customers                                636
# # CompetitionDistance                     1420
# # CompetitionOpenSinceMonth                 10
# # CompetitionOpenSinceYear                2012
# # Sales                                   6322
# # Name: 795018, dtype: object






import streamlit as st
import pandas as pd
import pickle

# Load your trained pipeline
with open(r'models/sales_prediction_pipeline.pkl', 'rb') as file:  # Use raw string or forward slashes
    model = pickle.load(file)

# Function to predict sales
def predict_sales(input_data):
    # Make predictions using the loaded model
    sales_prediction = model.predict(input_data)
    return sales_prediction

# Streamlit app
def main():
    st.title('Sales Prediction App')
    st.image("images/r1.jpg", caption="Rossmann")  # Use forward slashes for image path

    # Input widgets
    PromoInterval = st.selectbox("Promo Interval", ['No Promotion', 'Jan,Apr,Jul,Oct', 'Feb,May,Aug,Nov', 'Mar,Jun,Sept,Dec'])
    StoreType = st.radio("StoreType", ["Small Shop", "Medium Store", "Large Store", "Hypermarket"])
    Assortment = st.radio("Assortment", ["basic", "extra", "extended"])
    
    # Encode StateHoliday as 1 for 'Yes' and 0 for 'No'
    StateHoliday = st.radio("State Holiday", ["Yes", "No"])
    StateHoliday = 1 if StateHoliday == "Yes" else 0

    SchoolHoliday = st.radio("School Holiday", ["Yes", "No"])
    SchoolHoliday = 1 if SchoolHoliday == "Yes" else 0

    Promo = st.radio("Promotion", ["store is participating", "store is not participating"])
    Promo = 1 if Promo == "store is participating" else 0

    Store = st.slider("Store", 1, 1115)
    Customers = st.slider("Customers", 0, 7388)
    CompetitionDistance = st.slider("Competition Distance", 20, 75860)
    CompetitionOpenSinceMonth = st.slider("Competition Open Since Month", 1, 12)
    CompetitionOpenSinceYear = st.slider("Competition Open Since Year", 1998, 2015)

    # Store user inputs
    input_data = pd.DataFrame({
        'PromoInterval': [PromoInterval],
        'StoreType': [StoreType],
        'Assortment': [Assortment],
        'StateHoliday': [StateHoliday],
        'Store': [Store],
        'Customers': [Customers],
        'Promo': [Promo],
        'SchoolHoliday': [SchoolHoliday],
        'CompetitionDistance': [CompetitionDistance],
        'CompetitionOpenSinceMonth': [CompetitionOpenSinceMonth],
        'CompetitionOpenSinceYear': [CompetitionOpenSinceYear]
    })

    # Display input data
    st.subheader('Input Data:')
    st.write(input_data)

    # Predict sales
    if st.button('Predict Sales'):
        try:
            prediction = predict_sales(input_data)[0]  # Get the first prediction
            formatted_prediction = "{:.2f}".format(prediction)  # Format prediction to two decimal points
            st.write('Predicted Sales:', formatted_prediction)
        except Exception as e:
            st.error(f"An error occurred: {e}")


if __name__ == '__main__':
    main()