Added: Created visualizations that depict the traditional approach fo…

…r model assessment. Changed: Changed RMSE to CRMSE to remove ambiguity.
AAnzel · Jun 17, 2023 · 549fa8d · 549fa8d
1 parent 5c8a3af
commit 549fa8d
Show file tree

Hide file tree

Showing 2 changed files with 10,022 additions and 5,018 deletions.
diff --git a/Source/main.ipynb b/Source/main.ipynb
diff --git a/Source/polar_diagrams/src/polar_diagrams/polar_diagrams.py b/Source/polar_diagrams/src/polar_diagrams/polar_diagrams.py
@@ -104,14 +104,16 @@ def df_calculate_td_properties(df_input, string_reference_model,
  dict_result[string_one_model].append(math.degrees(
  math.acos(dict_result[string_one_model][-1])))
 
- # Calculating RMS
+ for string_one_model in list_all_features:
+ # Calculating CRMS
  dict_result[string_one_model].append(
- mean_squared_error(
+ math.sqrt(mean_squared_error(
  df_input[string_reference_model],
- df_input[string_one_model], squared=False))
+ df_input[string_one_model], squared=True) -
+ (np.mean(df_input[string_reference_model]) -
+ np.mean(df_input[string_one_model]))**2))
 
- for string_one_model in list_all_features:
- # Normalizing the RMS as in the paper
+ # Normalizing the CRMS as in the paper
  dict_result[string_one_model].append(
  dict_result[string_one_model][3] /
  dict_result[string_reference_model][0])
@@ -123,8 +125,8 @@ def df_calculate_td_properties(df_input, string_reference_model,
 
  df_result = pd.DataFrame().from_dict(
  dict_result, orient='index',
- columns=['Standard Deviation', 'Correlation', 'Angle', 'RMS',
- 'Normalized RMS', 'Normalized Standard Deviation'])
+ columns=['Standard Deviation', 'Correlation', 'Angle', 'CRMS',
+ 'Normalized CRMS', 'Normalized Standard Deviation'])
 
  df_result = df_result.reset_index().rename(columns={'index': 'Model'})
 
@@ -744,7 +746,7 @@ def _chart_create_diagram(list_df_input, string_reference_model,
  string_angular_column_label = 'Correlation'
  string_tooltip_label_1 = string_radial_column
  string_tooltip_label_2 = string_angular_column_label
- string_tooltip_label_3 = 'RMS'
+ string_tooltip_label_3 = 'CRMS'
  bool_only_half = True if (
  list_df_input[0][string_angular_column] <= 90).all() else False
  int_subplot_column_number = 1