Updated party colors & modified training params

de_analysis.py

@@ -13,12 +13,12 @@ grid_h = 11       # Grid height
 grid_w = 11       # Grid width
 radius = 2        # Neighbour radius
 step = 0.5        # Learning step
-ep = 300          # No of epochs
+ep = 500          # No of epochs
 
 # Load data
 dataset = ld.load_german_data()
 
-period_to_compass_year = {17:2005, 18:2013, 19:2017}
+period_to_compass_year = {17:2009, 18:2013, 19:2017}
 
 for period, df in dataset.items():
 
@@ -32,4 +32,3 @@ for period, df in dataset.items():
 
     # Predict and visualize output
     va.predict(model, data, grid_h, grid_w, de_name_color, pc.get_compass_parties(year=period_to_compass_year[period], country='de'))
-

uk_analysis.py

@@ -11,15 +11,12 @@ import numpy as np
 import pandas as pd
 import os
 
-
-
 # Train model
-grid_h = 30       # Grid height
-grid_w = 30       # Grid width
-radius = 3        # Neighbour radius
+grid_h = 13       # Grid height
+grid_w = 13       # Grid width
+radius = 2        # Neighbour radius
 step = 0.5
-ep = 1         # No of epochs
-
+ep = 300         # No of epochs
 
 period_to_compass_year = {'2015_uk':2015, '2017_uk':2017, '2019_uk':2019}
 main_directory = 'uk/csv'

voting_lib/party_colors.py

@@ -1,13 +1,13 @@
 de_name_color = {
     'AfD': 'blue', 
     'BÜ90/GR': 'green', 
-    'CDU/CSU': 'black', 
+    'CDU/CSU': 'orange', 
     'DIE LINKE.': 'purple', 
-    'FDP': 'yellow', 
+    'DIE LINKE': 'purple', 
+    'FDP': 'magenta', 
     'SPD': 'red',
 }
 
-
 uk_name_color = {
     'Conservative': 'blue', 
     'Democratic Unionist Party': 'salmon',
@@ -15,9 +15,11 @@ uk_name_color = {
     'Labour': 'red',
     'Liberal Democrat': 'darkorange', 
     'Plaid Cymru': 'darkgreen', 
-    'Scottish National Party': 'yellow', 
+    'Scottish National Party': 'magenta', 
     'Sinn Féin': 'yellowgreen', 
     'Social Democratic & Labour Party': 'cyan', 
     'UK Independence Party': 'purple',
     'Ulster Unionist Party': 'lightskyblue',
+    'Alba Party': 'black',
+    'Alliance': 'olive',
 }

voting_lib/voting_analysis.py

@@ -81,6 +81,7 @@ def predict(model, data, grid_h, grid_w, party_colors, comparison_data=pd.DataFr
        err = remove_NaN_rows_columns(normalize_df(part_distance_out) - normalize_df(comparison_data_dist))
        err = err * err
        plot_party_distances(err)
+       plt.title(f'MSE={np.nanmean(err.to_numpy()):.2f}')
        plt.show()
 
 def iter_neighbours(weights, hexagon=False):
@@ -218,7 +219,7 @@ def plot_parties(parties, party_colors, randomize_positions=False, new_plot=True
     
     for i, party in enumerate(party_index_mapping):
         print("Party ", party, " x = ", xs_disp[i], "y = ", ys_disp[i])
-        plt.scatter(xs_disp[i], ys_disp[i], label=party, zorder=2, c=colors[i], edgecolors='black')
+        plt.scatter(xs_disp[i], ys_disp[i], label=party, zorder=2, c=colors[i], edgecolors='None')
         
     plt.legend(title='Parties', bbox_to_anchor=(1.3, 1), loc='upper left')