First version release

app.py

@@ -179,6 +179,16 @@ def sales_growth(dataframe, fittedValues):
 
     return sales_growth
 
+@st.cache_data
+def merge_forecast_data(historical, test, future):
+    historical.rename(columns={historical.columns[0]: "Actual Sales"}, inplace=True)
+    test = test.to_frame()
+    test.rename(columns={test.columns[0]: "Predicted Sales"}, inplace=True)
+    future = future.to_frame()
+    future.rename(columns={future.columns[0]: "Future Forecasted Sales"}, inplace=True)
+    merged_dataframe = pd.merge(historical, test, future, on='Date', how='outer')
+    return merged_dataframe
+
 # TAPAS Model
 
 @st.cache_resource
@@ -222,7 +232,7 @@ def get_converted_answer(table, query):
 
 # Web Application
 st.title("Forecasting Dashboard 📈")
-st.subheader("Welcome User, start using the application by uploading your file in the sidebar!")
+st.subheader("Welcome User, start forecasting by uploading your file in the sidebar!")
 
 # Session States
 if 'uploaded' not in st.session_state:
@@ -233,26 +243,32 @@ if 'forecasted' not in st.session_state:
 
 # Sidebar Menu
 with st.sidebar:
-    st.title("SalesCast")
-    st.subheader("version 0.1")
-    uploaded_file = st.file_uploader("Upload your Store Data here (must atleast contain Date and Sales)", type=["csv"])
-    err = 0
+    # TODO Name for product
+    st.title("MaLaCast v1.0")
+    st.subheader("An intelligent sales forecasting system")
+    uploaded_file = st.file_uploader("Upload your store data here to proceed (must atleast contain Date and Sales)", type=["csv"])
     if uploaded_file is not None:
-      if uploaded_file.type != 'text/csv':
-            err = 1
-            st.info('Please upload in CSV format only...')
-      else: 
-        st.success("File uploaded successfully!")
-        df = pd.read_csv(uploaded_file, parse_dates=True)
-        st.write("Your uploaded data:")
-        st.write(df)
-
-        df = drop(df)
-        df = date_format(df)
-        merge_sort(df)
-        series = group_to_three(df)
-
-        st.session_state.uploaded = True
+      date_found, sales_found = False
+      for column in uploaded_file.columns:
+        if 'Date' in column:  
+          date_found = True
+        if 'Sales' in column:
+          sales_found = True
+        if(date_found and sales_found):
+          st.error('Please upload a csv containing both Date and Sales...')
+          st.stop()
+
+      st.success("File uploaded successfully!")
+      df = pd.read_csv(uploaded_file, parse_dates=True)
+      st.write("Your uploaded data:")
+      st.write(df)
+
+      df = drop(df)
+      df = date_format(df)
+      merge_sort(df)
+      series = group_to_three(df)
+
+      st.session_state.uploaded = True
 
     with open('sample.csv', 'rb') as f:
        st.download_button("Download our sample CSV", f, file_name='sample.csv')
@@ -288,21 +304,11 @@ if (st.session_state.uploaded):
     lower_series = pd.Series(confint[:, 0], index=index_of_fc)
     upper_series = pd.Series(confint[:, 1], index=index_of_fc)
 
-    # TODO Plot Test and Training
-    col1, col2 = st.columns(2)
-    with col1:
-       col1.header("Sales Forecast")
-       chart_data = pd.DataFrame(np.random.randn(20, 3), columns=["Forecasted", "Predicted", "Actual Sale"])
-       col1.line_chart(chart_data)
-      #  col1.line_chart(df['Sales'], x="Date", y="Actual Sales", color="#0000EE")
-    # plt.figure(figsize=(18,10))
     # plt.plot(df['Sales'], color='b', label = 'Actual Sales')
     # plt.plot(test_y, color='b')
     # plt.plot(fitted_series, color='r', label = 'Predicted Sales')
     # plt.title("SARIMAX - Forecast of Auto Business Retail Sales VS Actual Sales")
     # plt.legend(loc='upper left', fontsize=8)
-    # plt.show()
-    # plt.savefig('Auto Retail Future Sales Forecast 80-20.png')
 
     #Future predictions
     frequency = '3D'
@@ -329,9 +335,15 @@ if (st.session_state.uploaded):
     future_sales_growth = sales_growth(df, future_fitted_series)
     future_sales_growth = future_sales_growth.iloc[n_periods:]
     df = dates_df(future_sales_growth)
-    with col2:
-       col2.subheader("Forecasted sales in x days")
-       col2.write(df)
+    
+    col = st.columns(2)
+    with col[0]:
+       col[0].header("Sales Forecast")
+       merged_data = merge_forecast_data(df['Data'], fitted_series, future_fitted_series)
+       col[0].line_chart(merged_data)
+    with col[1]:
+       col[1].subheader("Forecasted sales in x days")
+       col[1].write(df)
     st.session_state.forecasted = True
 
     with st.form("question_form"):
@@ -340,7 +352,10 @@ if (st.session_state.uploaded):
 
     if query_button or question:
         answer = get_converted_answer(df, question)
-        st.subheader("The answer is:", answer)
+        if answer is not None:
+          st.subheader("The answer is:", answer)
+        else:
+           st.subheader("Answer is not found in table")
 
 # Hide Streamlit default style
 hide_st_style = """