Update app.py

app.py

@@ -5,15 +5,25 @@ from PIL import Image
 import requests
 from bokeh.plotting import figure
 from bokeh.models import HoverTool
-import pickle
+import joblib
+import os
 from date_features import getDateFeatures
 
-# Define path for the model and encoder from the pickle file
-model = pickle.load(open("model.pkl", "rb"))
-encoder = pickle.load(open("encoder.pkl", "rb"))
+
+# Get the current directory path
+current_dir = os.path.dirname(os.path.abspath(__file__))
+
+# Load the model from the pickle file
+model_path = os.path.join(current_dir, 'model.pkl')
+model = joblib.load(model_path)
+
+# Load the scaler from the pickle file
+encoder_path = os.path.join(current_dir, 'encoder.pkl')
+encoder = joblib.load(encoder_path)
+
 
 # Set Page Configurations
-st.set_page_config(page_title="ETA Prediction App", page_icon="fas fa-chart-line", layout="wide", initial_sidebar_state="auto")
+st.set_page_config(page_title="Sales Prediction App", page_icon="fas fa-chart-line", layout="wide", initial_sidebar_state="auto")
 
 # Loading GIF
 gif_url = "https://raw.githubusercontent.com/Gilbert-B/Forecasting-Sales/main/app/salesgif.gif"
@@ -24,24 +34,19 @@ menu = ['Home', 'About']
 choice = st.sidebar.selectbox("Select an option", menu)
 
 def predict(sales_data):
+    if sales_data.empty:
+        raise ValueError("No sales data provided.")
+    
+    # Perform the necessary data processing steps
     sales_data = getDateFeatures(sales_data).set_index('date')
-    # print(sales_data.columns)
-
-    # Make predictions for the next 8 weeks
-    prediction_inputs = []  # Initialize the list for prediction inputs
-
-    # Encode the prediction inputs
-    # numeric_columns = sales_data.select_dtypes(include=['int64', 'float64']).columns.tolist()
     numeric_columns = ['onpromotion', 'year', 'month', 'dayofmonth', 'dayofweek', 'dayofyear', 'weekofyear', 'quarter', 'year_weekofyear', 'sin(dayofyear)', 'cos(dayofyear)']
-    categoric_columns = ['store_id','category_id','city','store_type','cluster','holiday_type','is_holiday','is_month_start','is_month_end','is_quarter_start','is_quarter_end','is_year_start','is_year_end','is_weekend', 'season'] 
-    print(categoric_columns)
-    # encoder = BinaryEncoder(drop_invariant=False, return_df=True,)
-    # encoder.fit(sales_data[categoric_columns]) 
+    categoric_columns = ['store_id', 'category_id', 'city', 'store_type', 'cluster', 'holiday_type', 'is_holiday', 'is_month_start', 'is_month_end', 'is_quarter_start', 'is_quarter_end', 'is_year_start', 'is_year_end', 'is_weekend', 'season']
+    
     num = sales_data[numeric_columns]
     encoded_cat = encoder.transform(sales_data[categoric_columns])
     sales_data = pd.concat([num, encoded_cat], axis=1)
-    
-    # Make the prediction using the loaded machine learning model
+
+    # Make predictions using the pre-trained model
     predicted_sales = model.predict(sales_data)
 
     return predicted_sales
@@ -67,9 +72,12 @@ if choice == 'Home':
     categories = ['Category_' + str(i) for i in range(33)]
 
     with col1:
-        date = st.date_input("Date")
+        start_date = st.date_input("Start Date")
         # Convert the date to datetime format
-        date = pd.to_datetime(date)
+        start_date = pd.to_datetime(start_date)
+        end_date = st.date_input("End Date")
+        # Convert the date to datetime format
+        end_date = pd.to_datetime(end_date)
         onpromotion = st.number_input("How many products are on promotion?", min_value=0, step=1)
         selected_category = st.selectbox("Category", categories)
 
@@ -80,48 +88,48 @@ if choice == 'Home':
         selected_city = st.selectbox("City", cities)
         selected_cluster = st.selectbox("Cluster", clusters)
 
-    # Call getDateFeatures() function on sales_data (replace sales_data with your DataFrame)
-    sales_data = pd.DataFrame({
-        'date': [date],
-        'store_id': [selected_store],
-        'category_id': [selected_category],
-        'onpromotion': [onpromotion],
-        'city' :[selected_city],
-        'store_type': [selected_store1],
-        'cluster':[selected_cluster]
-        })
-    print(sales_data)
-    print(sales_data.info())
+    predicted_data = pd.DataFrame(columns=['Start Date', 'End Date', 'Store', 'Category', 'On Promotion', 'City', 'Cluster', 'Predicted Sales'])
 
 
     if st.button('Predict'):
-        sales = predict(sales_data)
-        formatted_sales = round(sales[0], 2)
-        st.write(f"Total sales for this week is: #{formatted_sales}")
-
+        if start_date > end_date:
+            st.error("Start date should be earlier than the end date.")
+        else:
+            with st.spinner('Predicting sales...'):
+                sales_data = pd.DataFrame({
+                    'date': pd.date_range(start=start_date, end=end_date),
+                    'store_id': [selected_store] * len(pd.date_range(start=start_date, end=end_date)),
+                    'category_id': [selected_category] * len(pd.date_range(start=start_date, end=end_date)),
+                    'onpromotion': [onpromotion] * len(pd.date_range(start=start_date, end=end_date)),
+                    'city': [selected_city] * len(pd.date_range(start=start_date, end=end_date)),
+                    'store_type': [selected_store1] * len(pd.date_range(start=start_date, end=end_date)),
+                    'cluster': [selected_cluster] * len(pd.date_range(start=start_date, end=end_date))
+                })
+                try:
+                    sales = predict(sales_data)
+                    formatted_sales = round(sales[0], 2)
+                    predicted_data = predicted_data.append({
+                         'Start Date': start_date,
+                         'End Date': end_date,
+                         'Store': selected_store,
+                         'Category': selected_category,
+                         'On Promotion': onpromotion,
+                         'City': selected_city,
+                         'Cluster': selected_cluster,
+                         'Predicted Sales': formatted_sales}, ignore_index=True)
+
+                    st.success(f"Total sales for the period is: #{formatted_sales}")
+                except ValueError as e:
+                    st.error(str(e))
 
-    # # Display the forecast results
-    # st.subheader("Sales Forecast for the Next 8 Weeks:")
-    # for week, sales in enumerate(predicted_sales, start=1):
-    #     st.write(f"Week {week}: {sales:.2f} units")
-
-    # # Update the line chart
-    # chart_data = pd.DataFrame({'Week': range(1, 9), 'Sales': predicted_sales})
-    # p = figure(plot_width=600, plot_height=400, title="Sales Forecast",
-    #            x_axis_label="Week", y_axis_label="Sales")
-
-    # p.line(chart_data['Week'], chart_data['Sales'], line_width=2)
-    # p.circle(chart_data['Week'], chart_data['Sales'], fill_color="white", size=6)
-    # p.add_tools(HoverTool(tooltips=[("Week", "@x"), ("Sales", "@y")]))
-    # st.bokeh_chart(p)
 
 # About section
 elif choice == 'About':
     # Load the banner image
     banner_image_url = "https://raw.githubusercontent.com/Gilbert-B/Forecasting-Sales/0d7b869515bysBoi5XxNGa3hayALLn9BK1VQqD69Dc/app/seer.png"
     banner_image = Image.open(requests.get(banner_image_url, stream=True).raw)
-
-    # Display the banner image
+ 
+   # Display the banner image
     st.image(banner_image, use_column_width=True)
     st.markdown('''
             <p style='font-size: 20px; font-style: italic;font-style: bold;'>
@@ -137,5 +145,6 @@ elif choice == 'About':
             and maximize their revenue potential in an ever-changing market landscape.
             </p>
             ''', unsafe_allow_html=True)
-    st.markdown("<p style='text-align: center;'>This Sales Forecasting App is developed using Streamlit and Python.</p>", unsafe_allow_html=True)
-    st.markdown("<p style='text-align: center;'>It demonstrates how machine learning can be used to predict sales for the next 8 weeks based on historical data.</p>", unsafe_allow_html=True)
+if st.button('Clear Data'):
+    predicted_data = pd.DataFrame(columns=['Start Date', 'End Date', 'Store', 'Category', 'On Promotion', 'City', 'Cluster', 'Predicted Sales'])
+    st.success("Data cleared successfully.")