Final Submission

prediction.py

@@ -1,47 +1,62 @@
 import numpy as np
 import pandas as pd
-import matplotlib.pyplot as plt
 from datetime import datetime, timedelta
 from sklearn.linear_model import LinearRegression
-from sklearn.preprocessing import StandardScaler
+from sklearn.preprocessing import MinMaxScaler
 import pickle
 import streamlit as st
+import matplotlib.pyplot as plt
 
-# Load All Files
-
+# Load all files
 with open('linreg_model.pkl', 'rb') as file_1:
-  model_lr = pickle.load(file_1)
+    model_lr = pickle.load(file_1)
 with open('num_columns.pkl', 'rb') as file_2:
-  scaler = pickle.load(file_2)
+    scaler = pickle.load(file_2)
+
+quantity_flow1 = pd.read_csv('quantity_flow1.csv')
+quantity_flow1['week_start_date'] = pd.to_datetime(quantity_flow1['week_start_date'])
+quantity_flow1.set_index('week_start_date', inplace=True)
 
+# Define the forecasting function
+def forecasting(month):
+    # :param month: how many months to predict
+
+    quantity_forecast = quantity_flow1.copy()
+    window = 2
+    for i in range(month):
+        X = np.array(quantity_forecast[-window:].values).reshape(1, -1)
+        X_scaled = scaler.transform(X)
+        new_idx = quantity_forecast.index[-1] + timedelta(weeks=1)
+        quantity_forecast.loc[new_idx] = round(model_lr.predict(X_scaled)[0])
+    return quantity_forecast.tail(month + window)
+
+# Define the Streamlit app
 def run():
-    # Load the time series data
-    quantity_flow1 = pd.read_csv('quantity_flow1.csv', index_col=0, parse_dates=True)
-    
-    # Function to Predict Quantity in Next n-weeks
-    def forecasting(month):
-        quantity_forecast = quantity_flow1.copy()
-        window = 2
-        for i in range(month):
-            X = np.array(quantity_forecast[-window:].values).reshape(1, -1)
-            X_scaled = scaler.transform(X)
-            new_idx = quantity_forecast.index[-1] + timedelta(weeks=1)
-            quantity_forecast[new_idx] = round(model_lr.predict(X_scaled)[0])
-        return quantity_forecast
+    st.title("Quantity Forecasting App")
     
-    # Set up the Streamlit app
-    st.title("Sales Forecasting")
-    
-    # Create a sidebar for user input
-    months = st.sidebar.slider("Select the number of weeks to forecast", 1, 24, 12)
-    
-    # Generate the forecast and plot the results
-    quantity_forecast = forecasting(months)
-    fig, ax = plt.subplots(figsize=(20, 5))
-    ax.plot(quantity_forecast, color='blue', label='forecast')
-    ax.plot(quantity_flow1, color='red', label='real')
-    ax.legend()
-    st.pyplot(fig)
-
-if __name__ == "__main__":
+    # Input the number of months to forecast
+    months_to_forecast = st.slider("How many months to forecast?", 1, 13, 3)
+
+    # Predict button
+    if st.button("Predict"):
+        # Get the forecast
+        quantity_forecast = forecasting(months_to_forecast)
+
+        # Display the forecast plot
+        fig, ax = plt.subplots(figsize=(20,5))
+        ax.plot(quantity_flow1, color='red', label='Real Quantity')
+        ax.plot(quantity_forecast, color='blue', label='Quantity Forecast')
+        ax.set_xlabel('Date')
+        ax.set_ylabel('Quantity')
+        ax.set_title(f"Quantity Forecast for the Next {months_to_forecast} Months")
+        ax.legend()
+        st.pyplot(fig)
+
+        # Reset the index of quantity_forecast and rename the index column to "Date"
+        quantity_forecast_table = quantity_forecast.reset_index().rename(columns={"index": "Date"})
+
+        # Display the forecasted sales in a table
+        st.write(quantity_forecast_table)
+
+if __name__ == '__main__':
     run()

quantity_flow1.csv

@@ -1,68 +1,68 @@
-quantity
-128808
-3839557
-4138934
-4149077
-4824273
-4008027
-4311010
-3882636
-4256541
-3883810
-5047898
-4303200
-4980761
-4620208
-4553719
-4921338
-5768747
-4684319
-1334134
-7172144
-5505616
-5394786
-5817233
-6260083
-5759699
-5709933
-5556481
-5275889
-5682207
-5824993
-5900392
-6143352
-6256332
-5021634
-5918497
-6112693
-5237531
-5554898
-5343014
-5912301
-4907074
-5269967
-4229764
-4878485
-4661839
-5234579
-4861820
-4465411
-4953197
-5717901
-5259200
-5768419
-4677997
-4111427
-4623965
-5265632
-4364421
-4703388
-4267946
-3731588
-4701262
-3732954
-4775638
-4787111
-4757590
-6001166
-4172070
+week_start_date,quantity
+2022-01-02,128808
+2022-01-09,3839557
+2022-01-16,4138934
+2022-01-23,4149077
+2022-01-30,4824273
+2022-02-06,4008027
+2022-02-13,4311010
+2022-02-20,3882636
+2022-02-27,4256541
+2022-03-06,3883810
+2022-03-13,5047898
+2022-03-20,4303200
+2022-03-27,4980761
+2022-04-03,4620208
+2022-04-10,4553719
+2022-04-17,4921338
+2022-04-24,5768747
+2022-05-01,4684319
+2022-05-08,1334134
+2022-05-15,7172144
+2022-05-22,5505616
+2022-05-29,5394786
+2022-06-05,5817233
+2022-06-12,6260083
+2022-06-19,5759699
+2022-06-26,5709933
+2022-07-03,5556481
+2022-07-10,5275889
+2022-07-17,5682207
+2022-07-24,5824993
+2022-07-31,5900392
+2022-08-07,6143352
+2022-08-14,6256332
+2022-08-21,5021634
+2022-08-28,5918497
+2022-09-04,6112693
+2022-09-11,5237531
+2022-09-18,5554898
+2022-09-25,5343014
+2022-10-02,5912301
+2022-10-09,4907074
+2022-10-16,5269967
+2022-10-23,4229764
+2022-10-30,4878485
+2022-11-06,4661839
+2022-11-13,5234579
+2022-11-20,4861820
+2022-11-27,4465411
+2022-12-04,4953197
+2022-12-11,5717901
+2022-12-18,5259200
+2022-12-25,5768419
+2023-01-01,4677997
+2023-01-08,4111427
+2023-01-15,4623965
+2023-01-22,5265632
+2023-01-29,4364421
+2023-02-05,4703388
+2023-02-12,4267946
+2023-02-19,3731588
+2023-02-26,4701262
+2023-03-05,3732954
+2023-03-12,4775638
+2023-03-19,4787111
+2023-03-26,4757590
+2023-04-02,6001166
+2023-04-09,4172070