m1

Streamlit_functions.py

@@ -1415,6 +1415,16 @@ def model_metrics_table_func():
     # model_metrics_df2 = model_metrics_df2.reset_index(drop = True)
     return model_metrics_df2
 
+def get_month_name(month_number):
+        
+        months = ["January", "February", "March", "April", "May", "June",
+          "July", "August", "September", "October", "November", "December"]
+        if 1 <= month_number <= 12:
+            return months[month_number - 1]  
+        else:
+            return "Invalid month number"
+    
+    
 
 def scenario_spend_forecasting(delta_df,start_date,end_date):
     
@@ -1453,30 +1463,104 @@ def scenario_spend_forecasting(delta_df,start_date,end_date):
     end_date = pd.to_datetime(end_date)
 
     cur_data = df[(df['Date'] >= start_date) & (df['Date'] <= end_date)]
-    cur_data = cur_data[spend_cols]
-    cur_data.columns = channels
-    data1 = pd.DataFrame(cur_data[channels].sum().transpose()).reset_index()
-    data1.columns = ["Channels","last_year_spends"]
-
-    df_modified = delta_df.merge(key_df,on = "Channel_name",how = "inner")
-    df_modified2 = df_modified.merge(data1,on = "Channels",how ="outer")
-    # df_modified2["Forecasted Spends"] =( df_modified2["last_year_spends"]*(1+df_modified2["Delta_percent"]/100)).astype(int)
-    df_modified2["Forecasted Spends"] =( df_modified2["last_year_spends"]*(1+df_modified2["Delta_percent"]/100)).apply(lambda x: "${:,.0f}".format(x))
-    df_modified2.index = df_modified2["Channels"]
-    df_modified2["Spend Change"] = (df_modified2["Delta_percent"]/100).apply(lambda x: "{:.0%}".format(x))
-    # df_modified2["Forecasted Spends"] = df_modified2["Forecasted Spends"].astype(int)
-    df_modified2["Last Year Spends"] = df_modified2["last_year_spends"].apply(lambda x: "${:,.0f}".format(x))
-    df_modified3 = df_modified2[["Last Year Spends","Forecasted Spends","Spend Change"]].transpose()
-    # df_modified2["forecasted_spends"] = 
-    # # df_modified = delta_percent
-    # # df_modified["Optimised Spends"] = df_modified["Current Spends"]*
-    df_modified3 = df_modified3[['BROADCAST TV', 'CABLE TV',
-       'CONNECTED & OTT TV', 'VIDEO', 'DISPLAY PROSPECTING',
-       'DISPLAY RETARGETING', 'SOCIAL PROSPECTING', 'SOCIAL RETARGETING',
-       'SEARCH BRAND', 'SEARCH NON-BRAND', 'DIGITAL PARTNERS', 'AUDIO',
-       'EMAIL']]
-
-    return df_modified3
+    cur_data["Month"] = cur_data["Date"].dt.month
+    # cur_data["Year"] = cur_data["Date"].dt.year
+    cur_data["Month year"] = cur_data["Month"].apply(get_month_name) + ' ' +(cur_data["Date"].dt.year+1).astype(str)
+    grp_cols = ['tv_broadcast_spend', 
+                'tv_cable_spend', 
+                'stream_video_spend', 
+                'olv_spend', 
+                'disp_prospect_spend', 
+                'disp_retarget_spend', 
+                'social_prospect_spend', 
+                'social_retarget_spend', 
+                'search_brand_spend', 
+                'search_nonbrand_spend', 
+                'cm_spend', 
+                'audio_spend', 
+                'email_spend',
+                "Month",
+                "Month year"]
+    
+    data2 = cur_data[grp_cols].groupby("Month year").sum()
+
+    data2.columns = [
+            'BROADCAST TV',
+            'CABLE TV',
+            'CONNECTED & OTT TV',
+            'VIDEO',
+            'DISPLAY PROSPECTING',
+            'DISPLAY RETARGETING',
+            'SOCIAL PROSPECTING',
+            'SOCIAL RETARGETING',
+            'SEARCH BRAND',
+            'SEARCH NON-BRAND',
+            'DIGITAL PARTNERS',
+            'AUDIO',
+            'EMAIL',
+            "Month"]
+    data2 = data2.sort_values("Month")
+    data2.drop(columns = ["Month"], inplace = True)
+
+    key_df = pd.DataFrame()
+    key_df["Channel_name"] = ["Email","DisplayRetargeting","\xa0Video","BroadcastTV","SocialRetargeting","Connected&OTTTV","SearchBrand","Audio","SocialProspecting","CableTV","DisplayProspecting","SearchNon-brand","DigitalPartners"]
+    key_df["Channels"] = ["EMAIL","DISPLAY RETARGETING","VIDEO","BROADCAST TV","SOCIAL RETARGETING","CONNECTED & OTT TV","SEARCH BRAND","AUDIO","SOCIAL PROSPECTING","CABLE TV","DISPLAY PROSPECTING","SEARCH NON-BRAND","DIGITAL PARTNERS"]
+    delta_df = delta_df.merge(key_df,on = "Channel_name",how = "inner")
+    print(delta_df)
+
+    data3 = data2.copy()
+    for channel in delta_df["Channels"]:
+        # print(channel)
+        delta_percent = delta_df[delta_df["Channels"]==channel]["Delta_percent"].iloc[0]
+        print(delta_percent)
+        data3[channel] = data3[channel]*(1+delta_percent/100)
+    print(data2)
+    print(data3)
+
+
+    ###### output dataframes
+    output_df2 = data3.copy()
+
+    #### percent change dataframe
+    delta_df2 = pd.DataFrame(data = delta_df["Delta_percent"].values,index = delta_df["Channels"])
+    print(delta_df2)
+    output_df1 = (pd.DataFrame(data2.sum()).transpose()).append(pd.DataFrame(data3.sum()).transpose()).append(delta_df2.transpose())
+    output_df1.index = ["Last Year Spends", "Forecasted Spends","Spends Change"]
+
+      
+
+    print(output_df1)
+
+
+
+
+    # print (data3)
+    # data3 = data2.append(key_df)
+    # print (data2)
+    # cur_data = cur_data[spend_cols]
+    # cur_data.columns = channels
+    # data1 = pd.DataFrame(cur_data[channels].sum().transpose()).reset_index()
+    # data1.columns = ["Channels","last_year_spends"]
+
+    # df_modified = delta_df.merge(key_df,on = "Channel_name",how = "inner")
+    # df_modified2 = df_modified.merge(data1,on = "Channels",how ="outer")
+    # # df_modified2["Forecasted Spends"] =( df_modified2["last_year_spends"]*(1+df_modified2["Delta_percent"]/100)).astype(int)
+    # df_modified2["Forecasted Spends"] =( df_modified2["last_year_spends"]*(1+df_modified2["Delta_percent"]/100)).apply(lambda x: "${:,.0f}".format(x))
+    # df_modified2.index = df_modified2["Channels"]
+    # df_modified2["Spend Change"] = (df_modified2["Delta_percent"]/100).apply(lambda x: "{:.0%}".format(x))
+    # # df_modified2["Forecasted Spends"] = df_modified2["Forecasted Spends"].astype(int)
+    # df_modified2["Last Year Spends"] = df_modified2["last_year_spends"].apply(lambda x: "${:,.0f}".format(x))
+    # df_modified3 = df_modified2[["Last Year Spends","Forecasted Spends","Spend Change"]].transpose()
+    # # df_modified2["forecasted_spends"] = 
+    # # # df_modified = delta_percent
+    # # # df_modified["Optimised Spends"] = df_modified["Current Spends"]*
+    # df_modified3 = df_modified3[['BROADCAST TV', 'CABLE TV',
+    #    'CONNECTED & OTT TV', 'VIDEO', 'DISPLAY PROSPECTING',
+    #    'DISPLAY RETARGETING', 'SOCIAL PROSPECTING', 'SOCIAL RETARGETING',
+    #    'SEARCH BRAND', 'SEARCH NON-BRAND', 'DIGITAL PARTNERS', 'AUDIO',
+    #    'EMAIL']]
+
+    return output_df1,output_df2
 
 def scenario_spend_forecasting2(delta_df,start_date,end_date):
     
@@ -1510,68 +1594,66 @@ def scenario_spend_forecasting2(delta_df,start_date,end_date):
     "DIGITAL PARTNERS"
 ]
 
-    import math    
-    start_date = pd.to_datetime(start_date)
-    end_date = pd.to_datetime(end_date)
+    # import math    
+    # start_date = pd.to_datetime(start_date)
+    # end_date = pd.to_datetime(end_date)
 
-    cur_data = df[(df['Date'] >= start_date) & (df['Date'] < end_date)]
-    cur_data = cur_data[spend_cols2]
-    cur_data.columns = channels2
+    # cur_data = df[(df['Date'] >= start_date) & (df['Date'] < end_date)]
+    # cur_data = cur_data[spend_cols2]
+    # cur_data.columns = channels2
     
-    cur_data["Date2"] = cur_data["Date"]+ pd.Timedelta(days=6)
-    cur_data["Month"] = cur_data["Date"].dt.month
+    # cur_data["Date2"] = cur_data["Date"]+ pd.Timedelta(days=6)
+    # cur_data["Month"] = cur_data["Date"].dt.month
 
     
-    # cur_data["Date"] = delta_df["Date"]
-    # cur_data["Date_diff"] = (cur_data["Date"]-start_date).dt.days
-    # cur_data["Date_diff_months"] =(np.ceil(cur_data["Date_diff"] / 30))
-
-    data2 = cur_data.groupby("Month").agg({
-        'Date':"min",
-        "Date2":"max",
-        'BROADCAST TV':"sum",
-        'CABLE TV':"sum",
-        'CONNECTED & OTT TV':"sum",
-        'VIDEO':"sum",
-        'DISPLAY PROSPECTING':"sum",
-        'DISPLAY RETARGETING':"sum",
-        'SOCIAL PROSPECTING':"sum",
-        'SOCIAL RETARGETING':"sum",
-        'SEARCH BRAND':"sum",
-        'SEARCH NON-BRAND':"sum",
-        'DIGITAL PARTNERS':"sum",
-        'AUDIO':"sum",
-        'EMAIL':"sum"
-    }).reset_index()
-
-    def get_month_name(month_number):
+    # # cur_data["Date"] = delta_df["Date"]
+    # # cur_data["Date_diff"] = (cur_data["Date"]-start_date).dt.days
+    # # cur_data["Date_diff_months"] =(np.ceil(cur_data["Date_diff"] / 30))
+
+    # data2 = cur_data.groupby("Month").agg({
+    #     'BROADCAST TV':"sum",
+    #     'CABLE TV':"sum",
+    #     'CONNECTED & OTT TV':"sum",
+    #     'VIDEO':"sum",
+    #     'DISPLAY PROSPECTING':"sum",
+    #     'DISPLAY RETARGETING':"sum",
+    #     'SOCIAL PROSPECTING':"sum",
+    #     'SOCIAL RETARGETING':"sum",
+    #     'SEARCH BRAND':"sum",
+    #     'SEARCH NON-BRAND':"sum",
+    #     'DIGITAL PARTNERS':"sum",
+    #     'AUDIO':"sum",
+    #     'EMAIL':"sum"
+    # }).reset_index()
+
+    # def get_month_name(month_number):
         
-        months = ["January", "February", "March", "April", "May", "June",
-          "July", "August", "September", "October", "November", "December"]
-        if 1 <= month_number <= 12:
-            return months[month_number - 1]  
-        else:
-            return "Invalid month number"
+    #     months = ["January", "February", "March", "April", "May", "June",
+    #       "July", "August", "September", "October", "November", "December"]
+    #     if 1 <= month_number <= 12:
+    #         return months[month_number - 1]  
+    #     else:
+    #         return "Invalid month number"
     
-    data2["Month year"] = data2["Month"].apply(get_month_name) + ' ' +(data2["Date"].dt.year+1).astype(str)   
-    # # print(data2.columns)
-    data2 = data2[['Month year' ,'BROADCAST TV', 'CABLE TV',
-       'CONNECTED & OTT TV', 'VIDEO', 'DISPLAY PROSPECTING',
-       'DISPLAY RETARGETING', 'SOCIAL PROSPECTING', 'SOCIAL RETARGETING',
-       'SEARCH BRAND', 'SEARCH NON-BRAND', 'DIGITAL PARTNERS', 'AUDIO',
-       'EMAIL']]
-    data2.columns = ['Month          ','BROADCAST TV', 'CABLE TV',
-       'CONNECTED & OTT TV', 'VIDEO', 'DISPLAY PROSPECTING',
-       'DISPLAY RETARGETING', 'SOCIAL PROSPECTING', 'SOCIAL RETARGETING',
-       'SEARCH BRAND', 'SEARCH NON-BRAND', 'DIGITAL PARTNERS', 'AUDIO',
-       'EMAIL']
+    # data2["Month year"] = data2["Month"].apply(get_month_name) + ' ' +(data2["Date"].dt.year+1).astype(str)   
+    # # # print(data2.columns)
+    # data2 = data2[['Month year' ,'BROADCAST TV', 'CABLE TV',
+    #    'CONNECTED & OTT TV', 'VIDEO', 'DISPLAY PROSPECTING',
+    #    'DISPLAY RETARGETING', 'SOCIAL PROSPECTING', 'SOCIAL RETARGETING',
+    #    'SEARCH BRAND', 'SEARCH NON-BRAND', 'DIGITAL PARTNERS', 'AUDIO',
+    #    'EMAIL']]
+    # data2.columns = ['Month          ','BROADCAST TV', 'CABLE TV',
+    #    'CONNECTED & OTT TV', 'VIDEO', 'DISPLAY PROSPECTING',
+    #    'DISPLAY RETARGETING', 'SOCIAL PROSPECTING', 'SOCIAL RETARGETING',
+    #    'SEARCH BRAND', 'SEARCH NON-BRAND', 'DIGITAL PARTNERS', 'AUDIO',
+    #    'EMAIL']
     
-    data2.set_index('Month          ', inplace=True)
-    for c in ['BROADCAST TV', 'CABLE TV',
-       'CONNECTED & OTT TV', 'VIDEO', 'DISPLAY PROSPECTING',
-       'DISPLAY RETARGETING', 'SOCIAL PROSPECTING', 'SOCIAL RETARGETING',
-       'SEARCH BRAND', 'SEARCH NON-BRAND', 'DIGITAL PARTNERS', 'AUDIO',
-       'EMAIL']:
-        data2[c] = data2[c].apply(lambda x: "${:,.0f}".format(x))
-    return data2
+    # data2.set_index('Month          ', inplace=True)
+    # for c in ['BROADCAST TV', 'CABLE TV',
+    #    'CONNECTED & OTT TV', 'VIDEO', 'DISPLAY PROSPECTING',
+    #    'DISPLAY RETARGETING', 'SOCIAL PROSPECTING', 'SOCIAL RETARGETING',
+    #    'SEARCH BRAND', 'SEARCH NON-BRAND', 'DIGITAL PARTNERS', 'AUDIO',
+    #    'EMAIL']:
+    #     data2[c] = data2[c].apply(lambda x: "${:,.0f}".format(x))
+    return key_df
 

pages/2_Scenario_Planner.py

@@ -2068,7 +2068,7 @@ if auth_status == True:
                 # print(a)
                 # print(summary_df_sorted['Actual_spend'].sum())
                 # print(summary_df_sorted['Actual_spend'])
-                print(col,summary_df_sorted)
+                # print(col,summary_df_sorted)
                 # print(summary_df_sorted['Old_sales'])
                 # print(col, "old efficiency ", a)
                 with bin_placeholder:
@@ -2218,18 +2218,20 @@ if auth_status == True:
             try:
                 
                 st.write("Forecasted Spends wrt. Channels ")
-                forecasted_table_df = sf.scenario_spend_forecasting(summary_df_sorted,start_date1-  relativedelta(years=1),end_date1-  relativedelta(years=1))
-                
-                forecasted_table_df2 = sf.scenario_spend_forecasting2(summary_df_sorted,start_date1-  relativedelta(years=1),end_date1-  relativedelta(years=1))
+                output_df1, output_df2 = sf.scenario_spend_forecasting(summary_df_sorted,start_date1-  relativedelta(years=1),end_date1-  relativedelta(years=1))
+                forecasted_table_df   = output_df1.copy()
+                forecasted_table_df.iloc[:2] = forecasted_table_df.iloc[:2].applymap(lambda x: "{:,.0f}".format(x))  
+                forecasted_table_df.iloc[-1] = forecasted_table_df.iloc[-1].apply(lambda x: "{:.1f}%".format(x))
                 st.dataframe(forecasted_table_df)
 
                 st.write("Monthly Breakdown Of Forecasted Spends wrt. Channels ")
+                forecasted_table_df2 = output_df2.applymap(lambda x: "{:,.0f}".format(x))
                 st.dataframe(forecasted_table_df2)
 
 
             except:
                 st.warning("Please make sure the base data is updated")
-
+            
             def save_report_forecast(forecasted_table_df,forecasted_table_df2):
                 
                 # Convert the DataFrame to an Excel file in memory
@@ -2252,7 +2254,7 @@ if auth_status == True:
             
             st.download_button(
                     "Download Report",
-                    data = save_report_forecast(forecasted_table_df,forecasted_table_df2),
+                    data = save_report_forecast(output_df1,output_df2),
                     file_name = report_name+".xlsx",
                     mime="application/vnd.ms-excel",
                     # on_click=lambda: save_report_forecast(forecasted_table_df,report_name),

summary_df.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cdae4dbff6029967edec8ee6ea14de7d682a0c536c7c0e303557ee644138c112
+oid sha256:89f9c3faa933ceebd7bee0c47ec71d04b7b8bdf401cd2ad55f0e22604203216f
 size 1822