m1

Model_Result_Overview.py

@@ -35,7 +35,7 @@ def get_random_effects(media_data, panel_col, mdf):
     random_eff_df = pd.DataFrame(columns=[panel_col, "random_effect"])
 
     for i, market in enumerate(media_data[panel_col].unique()):
-        print(i, end='\r')
+        # print(i, end='\r')
         intercept = mdf.random_effects[market].values[0]
         random_eff_df.loc[i, 'random_effect'] = intercept
         random_eff_df.loc[i, panel_col] = market
@@ -245,7 +245,7 @@ if auth_status:
 
     # with columns[0]:
     #     st.metric(label='Spends', value=format_numbers(float(scenario.actual_total_spends)))
-    # ###print(f"##################### {scenario.actual_total_sales} ##################")
+    # #### print(f"##################### {scenario.actual_total_sales} ##################")
     # with columns[1]:
     #     st.metric(label=target, value=format_numbers(float(scenario.actual_total_sales),include_indicator=False))
 

Streamlit_functions.py

@@ -348,8 +348,8 @@ def waterfall(start_date,end_date,btn_chart):
         font=dict(size=16),
         # align='left'
     )
-    # print(cur_data)
-    # print(prev_data)
+    # # print(cur_data)
+    # # print(prev_data)
     # fig.show()
     return fig
 
@@ -738,7 +738,7 @@ def media_decomp():
 
     media_cols = media_decomp_df.columns
     for i in range(2,len(media_cols)):
-    #     print(media_cols[i])
+    #     # print(media_cols[i])
         cumulative_df[media_cols[i]] = cumulative_df[media_cols[i]] + cumulative_df[media_cols[i-1]]
     # cumulative_df
 
@@ -1081,22 +1081,21 @@ def scenario_spend_forecasting2(delta_df,start_date,end_date):
     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 = 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["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))
+    # 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("Date_diff_months").agg({
+    data2 = cur_data.groupby("Month").agg({
         'Date':"min",
-        "Date2":"max"
-    }).reset_index()
-
-
-    data1 = cur_data.groupby("Date_diff_months").agg({
+        "Date2":"max",
         'BROADCAST TV':"sum",
         'CABLE TV':"sum",
         'CONNECTED & OTT TV':"sum",
@@ -1110,41 +1109,67 @@ def scenario_spend_forecasting2(delta_df,start_date,end_date):
         'DIGITAL PARTNERS':"sum",
         'AUDIO':"sum",
         'EMAIL':"sum"
-    }).transpose()
+    }).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_list = cur_data["Date_diff_months"].unique()
-    data1["Channels"]=data1.index
-    df_modified = delta_df.merge(key_df,on = "Channel_name",how = "inner")
-    df_modified2 = df_modified.merge(data1,on = "Channels",how ="outer")
-    df_modified2.index = df_modified2["Channels"]
+    data2["Month year"] = data2["Month"].apply(get_month_name) + ' ' +(data2["Date"].dt.year+1).astype(str)   
+    print(data2.columns)
+    data2 = data2[['Month year' ,'Date', 'Date2', '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          ','Base Data Start Date', 'Base Data End Date', '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['Base Data Start Date'] = data2['Base Data Start Date'].dt.date
+    data2['Base Data End Date'] = data2['Base Data End Date'].dt.date
+    #.transpose()
+    # st.dataframe(data2)
+    # st.dataframe(data1)
+    # months_list = cur_data["Month"].unique()
+    # data1["Channels"]=data1.index
+    # df_modified = delta_df.merge(key_df,on = "Channel_name",how = "inner")
+    # df_modified2 = df_modified.merge(data1,on = "Channels",how ="outer")
+    # df_modified2.index = df_modified2["Channels"]
 
-    data3 = pd.DataFrame(index = data1.index)
-    for c in months_list:
-        data3[c] = df_modified2[c]*(1+df_modified2["Delta_percent"]/100)
+    # data3 = pd.DataFrame(index = data1.index)
+    # for c in months_list:
+    #     data3[c] = df_modified2[c]*(1+df_modified2["Delta_percent"]/100)
     
-    df1 = df_modified2[months_list].transpose()
-    df1["Metrics"] = "Last Year Spends"
+    # df1 = df_modified2[months_list].transpose()
+    # df1["Metrics"] = "Last Year Spends"
     
-    data3 = data3.transpose()
-    data3 = data3.astype(int)
-    data2.index = data2["Date_diff_months"]
-    data2.columns = ["Date_diff_months","start date","end date"]
-    data3["start date"] = data2["start date"].dt.date
-    data3["end date"] = data2["end date"].dt.date
-    data3["Month"] = data3.index
-    cols = ["Month","start date","end date",'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']
+    # data3 = data3.transpose()
+    # data3 = data3.astype(int)
+    # # data2.index = data2["Date_diff_months"]
+    # data2.columns = ["start date","end date"]
+    # data3["start date"] = data2["start date"].dt.date
+    # data3["end date"] = data2["end date"].dt.date
+    # data3["Month"] = data3.index
+    # cols = ["Month","start date","end date",'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']
     # data3["Metrics"] = "Forecasted Year Spends"
 
     # df2 = df_modified2["Delta_percent"].transpose()
@@ -1187,6 +1212,6 @@ def scenario_spend_forecasting2(delta_df,start_date,end_date):
     # # # # df_modified = delta_percent
     # # # # df_modified["Optimised Spends"] = df_modified["Current Spends"]*
     # # spend_cols1 = pd.DataFrame(spend_cols)[0].to_list()
-    
-    return data3[cols]
+    data2.set_index('Month          ', inplace=True)
+    return data2
 

classes.py

@@ -4,7 +4,6 @@ from collections import OrderedDict
 import pandas as pd
 from numerize.numerize import numerize
 # from gekko import GEKKO
-
 def class_to_dict(class_instance):
     attr_dict = {}
     if isinstance(class_instance, Channel):
@@ -68,7 +67,7 @@ class Channel:
         self.dates = dates
         self.conversion_rate = conversion_rate
         self.actual_spends = spends.copy()
-        # self.actual_sales = sales.copy()
+        self.actual_sales = sales.copy()
 
         if modified_spends is None:
             self.modified_spends = self.actual_spends.copy()
@@ -84,8 +83,8 @@ class Channel:
 
         self.upper_limit = self.actual_spends.max() + self.actual_spends.std()
         self.power = np.ceil(np.log(self.actual_spends.max()) / np.log(10)) - 3
-        self.actual_sales = None
-        self.actual_sales = self.response_curve(self.actual_spends)#sales.copy()#
+        # self.actual_sales = None
+        # self.actual_sales = self.response_curve(self.actual_spends)#sales.copy()#
         self.actual_total_spends = self.actual_spends.sum()
         self.actual_total_sales = self.actual_sales.sum()
         self.modified_sales = self.calculate_sales()
@@ -106,7 +105,7 @@ class Channel:
         )
 
     def calculate_sales(self):
-        print("in calc_sales")
+        # # print("in calc_sales")
         return self.response_curve(self.modified_spends)
 
     def hill_equation(x, Kd, n):
@@ -119,27 +118,40 @@ class Channel:
                 self.upper_limit + (x - self.upper_limit) * self.upper_limit / x,
             )
         if self.response_curve_type == "hill-eq":
-            print("lalala")
-            print(self.name)
+            # dividing_parameter = check_dividing_parameter()
+            # # print("lalala")
+            # # print(self.name)
+            if len(x) == 1:
+                dividing_rate = 104
+                # x = np.sum(x)
+            else:
+                dividing_rate = 1
+                # x = np.sum(x)
+            # dividing_rate = 104
             Kd= self.response_curve_params["Kd"]
             n= self.response_curve_params["n"]
             x_min= self.response_curve_params["x_min"]
             x_max= self.response_curve_params["x_max"]
             y_min= self.response_curve_params["y_min"]
             y_max= self.response_curve_params['y_max']  
-            # print(x_min)
-            # print(Kd,n,x_min,x_max,y_min,y_max)
-            print(np.sum(x)/104)
-            x_inp = ( x- x_min) / (x_max - x_min)
-            # print(x_inp)
+            # # print(x_min)
+            # # print(Kd,n,x_min,x_max,y_min,y_max)
+            # # print(np.sum(x)/104)
+            x_inp = ( x/dividing_rate- x_min) / (x_max - x_min)
+            # # print("x",x)
+            # # print("x_inp",x_inp)
             x_out = x_inp**n / (Kd**n + x_inp**n) #self.hill_equation(x_inp,Kd, n)
-            print(x_out)
-            
-            sales = (y_max - y_min)*x_out + y_min
+            # # print("x_out",x_out)
+
+
+            x_val_inv = (x_out*x_max + (1 - x_out) * x_min)
+            sales = (x_val_inv*y_min/y_max)*dividing_rate
+            # sales = ((x_max - x_min)*x_out + x_min)*dividing_rate
+
             sales[np.isnan(sales)] = 0
-            print(sales) 
-            print(np.sum(sales))
-            # print(sales)
+            # # print(sales) 
+            # # print(np.sum(sales))
+            # # print("sales",sales)
         if self.response_curve_type == "s-curve":
             if self.power >= 0:
                 x = x / 10**self.power
@@ -254,14 +266,18 @@ class Scenario:
         return total_modified_spends
 
     def calculate_actual_total_sales(self):
-        total_actual_sales = self.constant.sum() + self.correction.sum()
+        total_actual_sales = 0 #self.constant.sum() + 
+        # # print(self.correction)
         for channel in self.channels.values():
             total_actual_sales += channel.actual_total_sales
+            # # print(channel.actual_total_sales)
+        # # print(total_actual_sales)
         return total_actual_sales
 
     def calculate_modified_total_sales(self):
-        total_modified_sales = self.constant.sum() + self.correction.sum()
+        total_modified_sales = 0 #self.constant.sum() + self.correction.sum()
         for channel in self.channels.values():
+            # print(channel,channel.modified_total_sales)
             total_modified_sales += channel.modified_total_sales
         return total_modified_sales
 
@@ -322,9 +338,9 @@ class Scenario:
     
     def cost_func(channel,x):
         x_inp = (x/104 - param_dicts["x_min"][channel]) / (param_dicts["x_max"][channel] - param_dicts["x_min"][channel])
-    #     print(x_inp)
+    #     # print(x_inp)
         x_out = hill_equation(x_inp, param_dicts["Kd"][channel], param_dicts["n"][channel])
-    #     print(x_out)
+    #     # print(x_out)
     #    
         return (param_dicts["y_max"][channel] - param_dicts["y_min"][channel])*(x_out + param_dicts["y_min"][channel])*104
     
@@ -332,7 +348,7 @@ class Scenario:
     
 
     def optimize_spends(self, sales_percent, channels_list, algo="trust-constr"):
-        print("%"*100)
+        # # print("%"*100)
         desired_sales = self.actual_total_sales * (1 + sales_percent / 100.0)
 
         def constraint(x):
@@ -346,10 +362,11 @@ class Scenario:
                 (1+np.array([-50.0, 100.0]) / 100.0)
                 * self.channels[ch].actual_total_spends
             )
-
+            # # print(self.channels[ch].actual_total_spends)
         initial_point = []
         for bound in bounds:
             initial_point.append(bound[0])
+            # initial_point = np.nan_to_num(initial_point, nan=0.0, posinf=0.0, neginf=0.0)
 
         power = np.ceil(np.log(sum(initial_point)) / np.log(10))
 
@@ -406,8 +423,8 @@ class Scenario:
                 self.update(channel_name, modified_spends)
             return -1 * self.modified_total_sales
 
-        print(bounds)
-        print("$"*100)
+        # # print(bounds)
+        # # print("$"*100)
         res = minimize(
             lambda x: objective_function(x)  / 1e3,
             method="trust-constr",
@@ -424,7 +441,7 @@ class Scenario:
         # bounds=bounds,
         # tol=1e-16
         # )
-        print(res)
+        # # print(res)
         for channel_name, modified_spends in zip(channels_list, res.x):
             self.update(channel_name, modified_spends)
 
@@ -440,9 +457,9 @@ class Scenario:
         param_dicts = {col: response_curve_params[col].to_dict() for col in response_curve_params.columns}
         
         x_inp = (x/104 - param_dicts["x_min"][channel]) / (param_dicts["x_max"][channel] - param_dicts["x_min"][channel])
-    #     print(x_inp)
+    #     # print(x_inp)
         x_out = self.hill_equation(x_inp, param_dicts["Kd"][channel], param_dicts["n"][channel])
-    #     print(x_out)
+    #     # print(x_out)
     #    
         return (param_dicts["y_max"][channel] - param_dicts["y_min"][channel])*(x_out + param_dicts["y_min"][channel])*104
         
@@ -462,7 +479,7 @@ class Scenario:
 
     #     x_vars=[]
     #     x_vars = [m.Var(value=param_dicts["current_spends"][_], lb=param_dicts["x_min"][_]*104, ub=5*param_dicts["current_spends"][_]) for _ in channels_list]
-    #     print(x_vars)
+    #     # print(x_vars)
     # #     x_vars,lower_bounds
         
     #     # Define the objective function to minimize
@@ -470,8 +487,8 @@ class Scenario:
     #     spends = 0 
     #     i = 0
     #     for i,c in enumerate(channels_list):
-    #         # print(c)
-    #         # print(x_vars[i])
+    #         # # print(c)
+    #         # # print(x_vars[i])
     #         cost = cost + (self.cost_func(c, x_vars[i]))
     #         spends = spends +x_vars[i]
 
@@ -486,7 +503,7 @@ class Scenario:
     #     m.solve(disp=True)
         
     #     for i, var in enumerate(x_vars):
-    #         print(f"x{i+1} = {var.value[0]}")
+    #         # print(f"x{i+1} = {var.value[0]}")
         
     #     for channel_name, modified_spends in zip(channels_list, x_vars):
     #         self.update(channel_name, modified_spends.value[0])

pages/3_Saved_Scenarios.py

@@ -32,7 +32,7 @@ def comparison_scenarios_df():
     summary_df_spend = None
     summary_df_prospect = None
     # summary_df_efficiency = None
-    #=print(scenarios_to_download)
+    #=# print(scenarios_to_download)
     for scenario_name in scenarios_to_compare:
         scenario_dict =  st.session_state['saved_scenarios'][scenario_name]
         _spends = []
@@ -295,7 +295,7 @@ def download_scenarios():
     wb.remove(wb.active)
     st.session_state['xlsx_buffer'] = io.BytesIO()
     summary_df = None
-    #print(scenarios_to_download)
+    ## print(scenarios_to_download)
     for scenario_name in scenarios_to_download:
         scenario_dict =  st.session_state['saved_scenarios'][scenario_name]
         _spends = []
@@ -385,8 +385,8 @@ auth_status = st.session_state.get('authentication_status')
 if auth_status == True:
     is_state_initiaized = st.session_state.get('initialized',False)
     if not is_state_initiaized:
-        #print("Scenario page state reloaded")
-        initialize_data()
+        ## print("Scenario page state reloaded")
+        initialize_data(target_file = "Overview_data_test_panel@#prospects.xlsx")
 
 
     saved_scenarios = st.session_state['saved_scenarios']
@@ -429,7 +429,8 @@ if auth_status == True:
             
         column_1, column_2,column_3 = st.columns((6,1,1))
         with column_1:
-            st.header(selected_scenario)
+            st.markdown(f'<span style="font-size:28px"><strong>Selected Scenario:</strong> {selected_scenario}</span>', unsafe_allow_html=True)
+            # st.header(f"Selected Scenario: {selected_scenario}")
         # with column_3:
         #     st.write("")
             # st.button('Delete scenario', on_click=delete_scenario)

response_curves_model_quality.py

@@ -112,7 +112,7 @@ def data_output(channel,X,y,y_fit_inv,x_ext_data,y_fit_inv_ext):
     for i in range(len(y_fit_inv_ext)):
         y_fit_inv_v2_ext.append(y_fit_inv_ext[i][0])
     
-#     print(x_ext_data)
+#     # print(x_ext_data)
     ext_df = pd.DataFrame()
     ext_df[f'{channel}_Spends'] = x_ext_data
     ext_df[fit_col] = y_fit_inv_v2_ext
@@ -125,7 +125,7 @@ def data_output(channel,X,y,y_fit_inv,x_ext_data,y_fit_inv_ext):
 
     ext_df['MAT'] = ["ext","ext","ext"]
     
-    # print(ext_df)
+    # # print(ext_df)
     plot_df= plot_df.append(ext_df)
     return plot_df
 
@@ -148,7 +148,7 @@ def input_data(df,spend_col,prospect_col):
     return X,y,x_data,y_data,x_minmax,y_minmax
 
 def extend_s_curve(x_max,x_minmax,y_minmax, Kd_fit, n_fit):
-    # print(x_max)
+    # # print(x_max)
     x_ext_data = [x_max*1.2,x_max*1.3,x_max*1.5]
 #     x_ext_data = [1500000,2000000,2500000]
 #     x_ext_data = [x_max+100,x_max+200,x_max+5000]
@@ -157,7 +157,7 @@ def extend_s_curve(x_max,x_minmax,y_minmax, Kd_fit, n_fit):
     for i in range(len(x_scaled)):
         x_data.append(x_scaled[i][0])
         
-    # print(x_data)
+    # # print(x_data)
     y_fit = hill_equation(x_data, Kd_fit, n_fit)
     y_fit_inv = y_minmax.inverse_transform(np.array(y_fit).reshape(-1,1))
     
@@ -170,8 +170,8 @@ def fit_data(spend_col,prospect_col,channel):
 
     X,y,x_data,y_data,x_minmax,y_minmax = input_data(temp_df,spend_col,prospect_col)
     y_fit, y_fit_inv, Kd_fit, n_fit = hill_func(x_data,y_data,x_minmax,y_minmax)
-    # print('k: ',Kd_fit)
-    # print('n: ', n_fit)
+    # # print('k: ',Kd_fit)
+    # # print('n: ', n_fit)
     
     ##### extend_s_curve
     x_ext_data,y_fit_inv_ext=  extend_s_curve(temp_df[spend_col].max(),x_minmax,y_minmax, Kd_fit, n_fit)
@@ -183,7 +183,7 @@ plotly_data = fit_data(spend_cols[0],prospect_cols[0],channel_cols[0])
 plotly_data.tail()
 
 for i in range(1,13):
-    # print(i)
+    # # print(i)
     pdf =  fit_data(spend_cols[i],prospect_cols[i],channel_cols[i])
     plotly_data = plotly_data.merge(pdf,on = ["Date","MAT"],how = "left")
     
@@ -210,6 +210,8 @@ def response_curves(channel,x_modified,y_modified):
     ))
     
     plotly_data2 = plotly_data.copy()
+    plotly_data2 = plotly_data[plotly_data[x_col].isnull()==False]
+    print(plotly_data[plotly_data2['Date'] == plotly_data2['Date'].max()][x_col])
     # .dropna(subset=[x_col]).reset_index(inplace = True)
     fig.add_trace(go.Scatter(
         x=plotly_data[plotly_data2['Date'] == plotly_data2['Date'].max()][x_col],
@@ -217,7 +219,7 @@ def response_curves(channel,x_modified,y_modified):
         mode='markers',
         marker=dict(
         size=13  # Adjust the size value to make the markers larger or smaller
-        , color = 'green'
+        , color = 'yellow'
         ),
         name="Current Spends"
     ))
@@ -356,7 +358,7 @@ def data_output(channel,X,y,y_fit_inv,x_ext_data,y_fit_inv_ext):
     for i in range(len(y_fit_inv_ext)):
         y_fit_inv_v2_ext.append(y_fit_inv_ext[i][0])
     
-#     print(x_ext_data)
+#     # print(x_ext_data)
     ext_df = pd.DataFrame()
     ext_df[f'{channel}_Spends'] = x_ext_data
     ext_df[fit_col] = y_fit_inv_v2_ext
@@ -369,7 +371,7 @@ def data_output(channel,X,y,y_fit_inv,x_ext_data,y_fit_inv_ext):
 
     ext_df['MAT'] = ["ext","ext","ext"]
     
-    # print(ext_df)
+    # # print(ext_df)
     plot_df= plot_df.append(ext_df)
     return plot_df
 
@@ -392,7 +394,7 @@ def input_data(df,spend_col,prospect_col):
     return X,y,x_data,y_data,x_minmax,y_minmax
 
 def extend_s_curve(x_max,x_minmax,y_minmax, Kd_fit, n_fit):
-    # print(x_max)
+    # # print(x_max)
     x_ext_data = [x_max*1.2,x_max*1.3,x_max*1.5]
 #     x_ext_data = [1500000,2000000,2500000]
 #     x_ext_data = [x_max+100,x_max+200,x_max+5000]
@@ -401,7 +403,7 @@ def extend_s_curve(x_max,x_minmax,y_minmax, Kd_fit, n_fit):
     for i in range(len(x_scaled)):
         x_data.append(x_scaled[i][0])
         
-    # print(x_data)
+    # # print(x_data)
     y_fit = hill_equation(x_data, Kd_fit, n_fit)
     y_fit_inv = y_minmax.inverse_transform(np.array(y_fit).reshape(-1,1))
     
@@ -414,8 +416,8 @@ def fit_data(spend_col,prospect_col,channel):
 
     X,y,x_data,y_data,x_minmax,y_minmax = input_data(temp_df,spend_col,prospect_col)
     y_fit, y_fit_inv, Kd_fit, n_fit = hill_func(x_data,y_data,x_minmax,y_minmax)
-    # print('k: ',Kd_fit)
-    # print('n: ', n_fit)
+    # # print('k: ',Kd_fit)
+    # # print('n: ', n_fit)
     
     ##### extend_s_curve
     x_ext_data,y_fit_inv_ext=  extend_s_curve(temp_df[spend_col].max(),x_minmax,y_minmax, Kd_fit, n_fit)
@@ -427,7 +429,7 @@ plotly_data = fit_data(spend_cols[0],prospect_cols[0],channel_cols[0])
 plotly_data.tail()
 
 for i in range(1,13):
-    # print(i)
+    # # print(i)
     pdf =  fit_data(spend_cols[i],prospect_cols[i],channel_cols[i])
     plotly_data = plotly_data.merge(pdf,on = ["Date","MAT"],how = "left")
     
@@ -455,10 +457,11 @@ def response_curves(channel,x_modified,y_modified):
     ))
     
     plotly_data2 = plotly_data.copy()
+    plotly_data2 = plotly_data[plotly_data[x_col].isnull()==False]
     # .dropna(subset=[x_col]).reset_index(inplace = True)
     fig.add_trace(go.Scatter(
-        x=plotly_data[plotly_data2['Date'] == plotly_data2['Date'].max()][x_col],
-        y=plotly_data[plotly_data2['Date'] == plotly_data2['Date'].max()][y_col],
+        x=plotly_data2[plotly_data2['Date'] == plotly_data2['Date'].max()][x_col],
+        y=plotly_data2[plotly_data2['Date'] == plotly_data2['Date'].max()][y_col],
         mode='markers',
         marker=dict(
         size=13  # Adjust the size value to make the markers larger or smaller

response_curves_model_quality_base.py

@@ -111,7 +111,7 @@ def data_output(channel,X,y,y_fit_inv,x_ext_data,y_fit_inv_ext):
     for i in range(len(y_fit_inv_ext)):
         y_fit_inv_v2_ext.append(y_fit_inv_ext[i][0])
     
-#     print(x_ext_data)
+#     # print(x_ext_data)
     ext_df = pd.DataFrame()
     ext_df[f'{channel}_Spends'] = x_ext_data
     ext_df[f'{channel}_Prospects'] = y_fit_inv_v2_ext
@@ -125,7 +125,7 @@ def data_output(channel,X,y,y_fit_inv,x_ext_data,y_fit_inv_ext):
 
     ext_df['MAT'] = ["ext","ext","ext"]
     
-    # print(ext_df.columns)
+    # # print(ext_df.columns)
     plot_df= plot_df.append(ext_df)
     return plot_df
 
@@ -148,7 +148,7 @@ def input_data(df,spend_col,prospect_col):
     return X,y,x_data,y_data,x_minmax,y_minmax
 
 def extend_s_curve(x_max,x_minmax,y_minmax, Kd_fit, n_fit):
-    # print(x_max)
+    # # print(x_max)
     x_ext_data = [x_max*1.2,x_max*1.3,x_max*1.5]
 #     x_ext_data = [1500000,2000000,2500000]
 #     x_ext_data = [x_max+100,x_max+200,x_max+5000]
@@ -157,7 +157,7 @@ def extend_s_curve(x_max,x_minmax,y_minmax, Kd_fit, n_fit):
     for i in range(len(x_scaled)):
         x_data.append(x_scaled[i][0])
         
-    # print(x_data)
+    # # print(x_data)
     y_fit = hill_equation(x_data, Kd_fit, n_fit)
     y_fit_inv = y_minmax.inverse_transform(np.array(y_fit).reshape(-1,1))
     
@@ -170,8 +170,8 @@ def fit_data(spend_col,prospect_col,channel):
 
     X,y,x_data,y_data,x_minmax,y_minmax = input_data(temp_df,spend_col,prospect_col)
     y_fit, y_fit_inv, Kd_fit, n_fit = hill_func(x_data,y_data,x_minmax,y_minmax)
-    # print('k: ',Kd_fit)
-    # print('n: ', n_fit)
+    # # print('k: ',Kd_fit)
+    # # print('n: ', n_fit)
     
     ##### extend_s_curve
     x_ext_data,y_fit_inv_ext=  extend_s_curve(temp_df[spend_col].max(),x_minmax,y_minmax, Kd_fit, n_fit)
@@ -183,7 +183,7 @@ plotly_data = fit_data(spend_cols[0],prospect_cols[0],channel_cols[0])
 plotly_data.tail()
 
 for i in range(1,13):
-    print(i)
+    # print(i)
     pdf =  fit_data(spend_cols[i],prospect_cols[i],channel_cols[i])
     plotly_data = plotly_data.merge(pdf,on = ["Date","MAT"],how = "left")
     

summary_df.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bbfe530ac30a045842611782bde814043862ff721a910f19cfbb8525f1dfb82a
+oid sha256:6828b9886464f55ba02eb0b7cd0f315e46f06c8ba520c041294ec6ff7d426965
 size 1822

utilities.py

@@ -206,12 +206,12 @@ def panel_level(input_df, date_column="Date"):
 
 
 def initialize_data(
-    panel=None, target_file="Overview_data_test_panel@#prospects.xlsx", updated_rcs=None, metrics=None
+    target_file, panel=None,  updated_rcs=None, metrics=None
 ):
     # uopx_conv_rates = {'streaming_impressions' : 0.007,'digital_impressions' : 0.007,'search_clicks' : 0.00719,'tv_impressions' : 0.000173,
     #                    "digital_clicks":0.005,"streaming_clicks":0.004,'streaming_spends':1,"tv_spends":1,"search_spends":1,
     #                    "digital_spends":1}
-    # print('State initialized')
+    # # print('State initialized')
 
     excel = pd.read_excel(target_file, sheet_name=None)
 
@@ -305,15 +305,16 @@ def initialize_data(
         if updated_rcs is not None and updated_rcs_key in list(updated_rcs.keys()):
             response_curves[inp_col] = updated_rcs[updated_rcs_key]
 
+        # # print(response_curves)
         ## conversion rates
         spend_col = [
             _col
             for _col in spend_df.columns
             if _col.startswith(inp_col.rsplit("_", 1)[0])
         ][0]
-
-        # print('#printing spendssss')
-        # print(spend_col)
+        # # print(spend_col)
+        # # print('## printing spendssss')
+        # # print(spend_col)
         conv = (
             spend_df.set_index("Week")[spend_col]
             / input_df.set_index("Date")[inp_col].clip(lower=1)
@@ -323,9 +324,10 @@ def initialize_data(
         conv_rates[inp_col] = list(conv.drop("Week", axis=1).mean().to_dict().values())[
             0
         ]
-        ##print('Before',conv_rates[inp_col])
+        # # print(conv_rates)
+        ### print('Before',conv_rates[inp_col])
         # conv_rates[inp_col] = uopx_conv_rates[inp_col]
-        ##print('After',(conv_rates[inp_col]))
+        ### print('After',(conv_rates[inp_col]))
 
         channel = Channel(
             name=inp_col,
@@ -352,12 +354,15 @@ def initialize_data(
             sales = channel.actual_sales
         else:
             sales += channel.actual_sales
+    # # print(actual_output_dic)
     other_contributions = (
         output_df.drop([*output_cols], axis=1).sum(axis=1, numeric_only=True).values
     )
     correction = output_df.drop("Date", axis=1).sum(axis=1).values - (
         sales + other_contributions
     )
+    # # print(other_contributions)
+    # # print(correction)
     scenario = Scenario(
         name="default",
         channels=channels,
@@ -400,6 +405,10 @@ def initialize_data(
         channel_name: False for channel_name in channel_list
     }
     st.session_state["disable_download_button"] = True
+    # if target_file == :
+    #     st.session_state["dividing_parameter"] =  
+    # else :
+
 
 
 def create_channel_summary(scenario):
@@ -700,7 +709,7 @@ def create_channel_spends_sales_plot(channel):
         df = raw_df.sort_values(by="Date")
         x = df.Date
         scenario = class_from_dict(st.session_state["default_scenario_dict"])
-        _sales = scenario.constant + scenario.correction
+        _sales = 0 #scenario.constant + scenario.correction
         channel_sales_spends_fig = make_subplots(specs=[[{"secondary_y": True}]])
         channel_sales_spends_fig.add_trace(
             go.Bar(

utilities_with_panel.py

@@ -98,7 +98,7 @@ DATA_PATH = './data'
 
 IMAGES_PATH = './data/images_224_224'
 
-# New - Sprint 2
+# New - S# print 2
 if 'bin_dict' not in st.session_state:
     
     with open("data_import.pkl", "rb") as f:
@@ -395,7 +395,7 @@ def initialize_data(target_col,selected_markets):
     # uopx_conv_rates = {'streaming_impressions' : 0.007,'digital_impressions' : 0.007,'search_clicks' : 0.00719,'tv_impressions' : 0.000173,
     #                    "digital_clicks":0.005,"streaming_clicks":0.004,'streaming_spends':1,"tv_spends":1,"search_spends":1,
     #                    "digital_spends":1}
-    #print('State initialized')
+    ## print('State initialized')
     # excel = pd.read_excel("data_test_overview_panel.xlsx",sheet_name=None)
     #excel = pd.read_excel("Overview_data_test_panel@#revenue.xlsx" + target_col + ".xlsx",sheet_name=None)
     
@@ -469,7 +469,7 @@ def initialize_data(target_col,selected_markets):
     for inp_col in channel_list:
         #st.write(inp_col)
 
-        # # New - Sprint 2
+        # # New - S# print 2
         # if is_panel:
         #     input_df1 = input_df.groupby([date_col]).agg({inp_col:'sum'}).reset_index() # aggregate spends on date
         #     spends = input_df1[inp_col].values
@@ -484,7 +484,7 @@ def initialize_data(target_col,selected_markets):
 
 
         # contribution
-        # New - Sprint 2
+        # New - S# print 2
         out_col = [_col for _col in output_df.columns if _col.startswith(inp_col)][0]
         if is_panel :
             output_df1 = output_df.groupby([date_col]).agg({out_col:'sum'}).reset_index()
@@ -505,12 +505,12 @@ def initialize_data(target_col,selected_markets):
             
         x = x.astype('float64')
         y = y.astype('float64')
-        #print('#printing yyyyyyyyy')
-        #print(inp_col)
-        #print(x.max())
-        #print(y.max())
+        ## print('## printing yyyyyyyyy')
+        ## print(inp_col)
+        ## print(x.max())
+        ## print(y.max())
         # st.write(y.max(),x.max())
-        print(y.max(),x.max())
+        # print(y.max(),x.max())
         if y.max()<=0.01:
             if x.max()<=0.01 :
                 st.write("here-here")
@@ -539,15 +539,15 @@ def initialize_data(target_col,selected_markets):
         ## conversion rates
         spend_col = [_col for _col in spend_df.columns if _col.startswith(inp_col.rsplit('_',1)[0])][0]
 
-        #print('#printing spendssss')
-        #print(spend_col)
+        ## print('## printing spendssss')
+        ## print(spend_col)
         conv = (spend_df.set_index('Week')[spend_col] / input_df.set_index('Date')[inp_col].clip(lower=1)).reset_index()
         conv.rename(columns={'index':'Week'},inplace=True)
         conv['year'] = conv.Week.dt.year
         conv_rates[inp_col] = list(conv.drop('Week',axis=1).mean().to_dict().values())[0]
-        ##print('Before',conv_rates[inp_col])
+        ### print('Before',conv_rates[inp_col])
         # conv_rates[inp_col] = uopx_conv_rates[inp_col]
-        ##print('After',(conv_rates[inp_col]))
+        ### print('After',(conv_rates[inp_col]))
         
         
         channel = Channel(name=inp_col,dates=dates,
@@ -617,7 +617,7 @@ def initialize_data(target_col,selected_markets):
 #     channel_list = []
 #     for col in raw_df.columns:
 #         if 'click' in col.lower() or 'spend' in col.lower() or 'imp' in col.lower():
-#             ##print(col)
+#             ### print(col)
 #             channel_list.append(col)
 #         else:
 #             pass
@@ -708,8 +708,8 @@ def create_channel_summary(scenario):
         if name_mod.lower().endswith(' imp'):
             name_mod = name_mod.replace('Imp', ' Impressions')
 
-        print(name_mod, channel.actual_total_spends, channel.conversion_rate,
-              channel.actual_total_spends * channel.conversion_rate)
+        # print(name_mod, channel.actual_total_spends, channel.conversion_rate,
+        channel.actual_total_spends * channel.conversion_rate
 
         summary_columns.append(name_mod)