changes in displaying analysis

Scripts/Analysis_final_mult.py

@@ -131,7 +131,7 @@ def plot_confusion_matrix(cm, classes,normalize=True,title='Confusion matrix',cm
    
    fmt = '.2f' if normalize else 'd'
    
-   thresh = cm.max() / 2
+   thresh = cm.max() / 1.3
 
    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):
        ax.text(j, i, format(cm_old[i, j], fmt),
@@ -756,7 +756,39 @@ def plotter(subsets, subana, phases, labels):
                                               gen_pred_auth,
                                                  average='weighted').round(3)
                            }
-                        
+
+                    gend = [comp]
+                    ind2 = ['author by gender']
+
+                    gend_an = {'accuracy': accuracy_score(df_dic[st][an]['df'].loc[df_dic[st][an]['df'][an+'_'+str(st)+'_SVM_gender'] == 'male'][an+'_'+str(st)+'_SVM_gender'],
+                                              df_dic[st][an]['df'].loc[df_dic[st][an]['df'][an+'_'+str(st)+'_SVM_gender'] == 'male'][an+'_'+str(st)+'_gender_pred_auth']).round(3),
+                        'precision': precision_score(df_dic[st][an]['df'][an+'_'+str(st)+'_SVM_gender'],
+                                              gen_pred_auth, pos_label = 'male',
+                                                 average='binary').round(3),
+                        'recall': recall_score(df_dic[st][an]['df'][an+'_'+str(st)+'_SVM_gender'],
+                                              gen_pred_auth, pos_label = 'male',
+                                                 average='binary').round(3),
+                        'f1-score': f1_score(df_dic[st][an]['df'][an+'_'+str(st)+'_SVM_gender'],
+                                              gen_pred_auth, pos_label = 'male',
+                                                 average='binary').round(3)
+                           }
+                    gend.append(gend_an)
+                    ind2.append('author by gender\n male')
+
+                    gend_an = {'accuracy': accuracy_score(df_dic[st][an]['df'].loc[df_dic[st][an]['df'][an+'_'+str(st)+'_SVM_gender'] == 'female'][an+'_'+str(st)+'_SVM_gender'],
+                                              df_dic[st][an]['df'].loc[df_dic[st][an]['df'][an+'_'+str(st)+'_SVM_gender'] == 'female'][an+'_'+str(st)+'_gender_pred_auth']).round(3),
+                        'precision': precision_score(df_dic[st][an]['df'][an+'_'+str(st)+'_SVM_gender'],
+                                              gen_pred_auth, pos_label = 'female',
+                                                 average='binary').round(3),
+                        'recall': recall_score(df_dic[st][an]['df'][an+'_'+str(st)+'_SVM_gender'],
+                                              gen_pred_auth, pos_label = 'female',
+                                                 average='binary').round(3),
+                        'f1-score': f1_score(df_dic[st][an]['df'][an+'_'+str(st)+'_SVM_gender'],
+                                              gen_pred_auth, pos_label = 'female',
+                                                 average='binary').round(3)
+                           }
+                    ind2.append('author by gender\n female')
+                    gend.append(gend_an)
                     tmp.append(comp)
                     index.append('author by gender')
                         
@@ -823,6 +855,20 @@ def plotter(subsets, subana, phases, labels):
             plt.tight_layout()
             plt.savefig(savedir+'barplots/overall_scores_{}_{}.pdf'.format(st, an))
             plt.savefig(savedir+'barplots/overall_scores_{}_{}.png'.format(st, an))
+
+
+            f= plt.figure(figsize=(10,5))
+            tmp_df =pd.DataFrame(gend, index = ind2)
+            tmp_df.plot(kind='barh', colormap = cmap, ax=f.gca())
+            plt.legend(loc='center left', bbox_to_anchor=(1.0, 0.5))
+            plt.xlabel(xlabel='Evaluation Measures for Different Subsets',fontsize ='large', fontweight='roman')
+            ax = plt.gca()
+            plt.setp(ax.get_xticklabels(), rotation=45, ha="right",rotation_mode="anchor")
+            plt.grid(True, axis ='y')
+            plt.ylim((0,0.7))
+            plt.tight_layout()
+            plt.savefig(savedir+'barplots/author_gender_scores_{}_{}.pdf'.format(st, an))
+            plt.savefig(savedir+'barplots/author_gender_scores_{}_{}.png'.format(st, an))