DDESONN — Plot Controls — Scenario 1 & 2 — Single Run: Scenario A

Plot Interface — Scenario 1 vs Scenario 2

This vignette demonstrates two supported plot configuration interfaces in DDESONN.

Scenario 1 — Independent Call Sites (training_overrides)

Scenario 1 configures plotting through three independent call sites inside training_overrides:

  1. training_overrides$per_epoch_plots
  2. training_overrides$performance_relevance
  3. training_overrides$evaluate_predictions_report_plots

Each plot category is configured separately.
This represents the minimal integration bridge interface.

For the unified umbrella interface (recommended), see:

plot-controls_scenario1-2_single-run_scenarioA.Rmd

Scenario 2 — Unified Umbrella (plot_controls)

Scenario 2 configures all plots through a single unified interface:

All plot categories (e.g., per_epoch, performance_relevance, evaluate_report) are defined under this one umbrella.

For comparison with the three-call-site structure (Scenario 1), see:

plot-controls_scenario1-2_single-run_scenarioA.Rmd

[PR] RESOLVED pr_cfg for update_performance_and_relevance (ensemble_number=0, do_ensemble=FALSE) list(saveEnabled = TRUE, viewAllPlots = FALSE, performance_high_mean_plots = TRUE, performance_low_mean_plots = TRUE, relevance_high_mean_plots = TRUE, relevance_low_mean_plots = TRUE, verbose = TRUE) [PR] knitr.in.progress=TRUE [PR] interactive()=FALSE [PR] option(DDESONN_OUTPUT_ROOT)=C:6f9Ge/ddesonn-vig-s1-s2/outputs/DDESONN_plots_scenarioA_s1 [PR] env(DDESONN_ARTIFACTS_ROOT)=C:6f9Ge/ddesonn-vig-s1-s2/outputs/DDESONN_plots_scenarioA_s1 [PR] run_id (head): Ensemble: 0 Model: 1 (n=1) [PR] pr_saveEnabled=TRUE [PR] pr_viewAllPlots=FALSE [PR] flags resolved (post viewAll): perf_high=TRUE perf_low=TRUE relev_high=TRUE relev_low=TRUE [PR TRACE] pr_viewAllPlots= FALSE pr_saveEnabled= TRUE [PR TRACE] rows: perf_high= 7 perf_low= 36 relev_high= 0 relev_low= 3 [PR TRACE] print_plots=TRUE [PR TRACE] CALL high(perf) [PR] performance_high_mean_plots class=list length=7 names(head)=confusion_matrix_TP,confusion_matrix_FP,confusion_matrix_TN,accuracy_precision_recall_f1_tuned_confusion_matrix_TP,accuracy_precision_recall_f1_tuned_confusion_matrix_TN [PR TRACE] CALL low(perf) [PR] performance_low_mean_plots class=list length=36 names(head)=quantization_error,topographic_error,clustering_quality_db,MSE,MAE [PR TRACE] CALL high(relev) [PR] relevance_high_mean_plots class=list length=0 [PR TRACE] CALL low(relev) [PR] relevance_low_mean_plots class=list length=3 names(head)=ndcg,diversity,serendipity


# ================================================================================
# ================================= CORE METRICS =================================
# ================================================================================

===== FINAL SUMMARY =====
Best epoch          : 167
Train accuracy      : 0.990000
Val accuracy        : 0.984000
Train loss          : 0.008937
Val loss            : 0.013941
Threshold           : 0.500000
Test accuracy       : 0.988000
Test loss           : 0.058435

===== TRAIN =====

Classification Report
precision recall f1-score support
0 0.996275 0.997927 0.997100 2412.000000
1 0.995387 0.991728 0.993554 1088.000000
accuracy 0.996000 0.996000 0.996000 3500.000000
macro avg 0.995831 0.994827 0.995327 3500.000000
weighted avg 0.995999 0.996000 0.995998 3500.000000
Confusion Matrix
Positive (1) Negative (0)
Positive (1) 1079 5
Negative (0) 9 2407

AUC/AUPRC AUC (ROC): 0.999922 AUPRC: 0.999831

===== VALIDATION =====

Classification Report
precision recall f1-score support
0 0.991718 0.977551 0.984584 490.000000
1 0.958801 0.984615 0.971537 260.000000
accuracy 0.980000 0.980000 0.980000 750.000000
macro avg 0.975260 0.981083 0.978060 750.000000
weighted avg 0.980307 0.980000 0.980061 750.000000
Confusion Matrix
Positive (1) Negative (0)
Positive (1) 256 11
Negative (0) 4 479

AUC/AUPRC AUC (ROC): 0.992708 AUPRC: 0.986047

===== TEST =====

Classification Report
precision recall f1-score support
0 0.998088 0.984906 0.991453 530.000000
1 0.964758 0.995455 0.979866 220.000000
accuracy 0.988000 0.988000 0.988000 750.000000
macro avg 0.981423 0.990180 0.985659 750.000000
weighted avg 0.988311 0.988000 0.988054 750.000000
Confusion Matrix
Positive (1) Negative (0)
Positive (1) 219 8
Negative (0) 1 522

AUC/AUPRC AUC (ROC): 0.996964 AUPRC: 0.996219

Scenario 1 — Saved plots





















































































































































































































Scenario 2 — One plot_controls umbrella (Single Run: Scenario A)

options(DDESONN_OUTPUT_ROOT = if (is_check_env) .vig_tmp_root else out2)
Sys.setenv(DDESONN_ARTIFACTS_ROOT = if (is_check_env) .vig_tmp_root else out2)

res_scenarioA_s2 <- ddesonn_run(
  x = x_train,
  y = y_train,
  classification_mode = "binary",
  hidden_sizes = c(64, 32),
  seeds = 1L,
  do_ensemble = FALSE,
  validation = list(x = x_valid, y = y_valid),
  test       = list(x = x_test,  y = y_test),
  training_overrides = list(
    init_method = "he",
    optimizer = "adagrad",
    lr = 0.125,
    lambda = 0.00028,
    activation_functions = list(relu, relu, sigmoid),
    dropout_rates = list(0.10),
    loss_type = "CrossEntropy",
    validation_metrics = TRUE,
    num_epochs = 1,
    final_summary_decimals = 6L
  ),
  plot_controls = list(
    per_epoch = list(
      saveEnabled = !is_check_env,
      loss_curve = TRUE,
      probe_plots = TRUE,
      verbose = TRUE
    ),

    # ============================================================
    #  SECTION: performance_relevance (Scenario 2)
    # ============================================================
    performance_relevance = list(                                                       
      saveEnabled = !is_check_env,                                                               
      viewAllPlots = FALSE,                                                             
      performance_high_mean_plots = TRUE,                                                
      performance_low_mean_plots  = TRUE,                                                
      relevance_high_mean_plots   = TRUE,                                                
      relevance_low_mean_plots    = TRUE,                                                
      verbose = TRUE                                                                    
    ),                                                                                  

    evaluate_report = list(
      accuracy_plot = TRUE,
      accuracy_plot_mode = "both",
      plot_roc = TRUE,
      plot_pr  = TRUE,
      show_auprc = TRUE,
      viewAllPlots = FALSE,
      verbose = TRUE
    )
  )
)
## [PR] RESOLVED pr_cfg for update_performance_and_relevance (ensemble_number=0, do_ensemble=FALSE)
## list(saveEnabled = TRUE, viewAllPlots = FALSE, performance_high_mean_plots = TRUE, 
##     performance_low_mean_plots = TRUE, relevance_high_mean_plots = TRUE, 
##     relevance_low_mean_plots = TRUE, verbose = TRUE) 
## [PR] knitr.in.progress=TRUE
## [PR] interactive()=FALSE
## [PR] option(DDESONN_OUTPUT_ROOT)=C:\Users\wfky1\AppData\Local\Temp\RtmpK6f9Ge/ddesonn-vig-s1-s2/outputs/DDESONN_plots_scenarioA_s2
## [PR] env(DDESONN_ARTIFACTS_ROOT)=C:\Users\wfky1\AppData\Local\Temp\RtmpK6f9Ge/ddesonn-vig-s1-s2/outputs/DDESONN_plots_scenarioA_s2
## [PR] run_id (head): Ensemble: 0 Model: 1 (n=1)
## [PR] pr_saveEnabled=TRUE
## [PR] pr_viewAllPlots=FALSE
## [PR] flags resolved (post viewAll): perf_high=TRUE perf_low=TRUE relev_high=TRUE relev_low=TRUE
## [PR TRACE] pr_viewAllPlots= FALSE  pr_saveEnabled= TRUE 
## [PR TRACE] rows: perf_high= 9  perf_low= 18  relev_high= 0  relev_low= 3 
## [PR TRACE] print_plots=TRUE
## [PR TRACE] CALL high(perf)
## [PR] performance_high_mean_plots class=list length=9 names(head)=confusion_matrix_TP,confusion_matrix_FP,confusion_matrix_TN,confusion_matrix_FN,accuracy_precision_recall_f1_tuned_confusion_matrix_TP
## [PR TRACE] CALL low(perf)
## [PR] performance_low_mean_plots class=list length=18 names(head)=quantization_error,topographic_error,clustering_quality_db,MSE,MAE
## [PR TRACE] CALL high(relev)
## [PR] relevance_high_mean_plots class=list length=0
## [PR TRACE] CALL low(relev)
## [PR] relevance_low_mean_plots class=list length=3 names(head)=ndcg,diversity,serendipity
## 
## ```text
## 
## # ================================================================================
## # ================================= CORE METRICS =================================
## # ================================================================================
## 
## ===== FINAL SUMMARY =====
## Best epoch          : 1
## Train accuracy      : 0.493429
## Val accuracy        : 0.697333
## Train loss          : 0.253650
## Val loss            : 0.206287
## Threshold           : 0.650000
## Test accuracy       : 0.801333
## Test loss           : 0.628412 
## ```
## 
## ===== TRAIN =====
## 
## Classification Report
## <table>
##  <thead>
##   <tr>
##    <th style="text-align:left;">   </th>
##    <th style="text-align:left;"> precision </th>
##    <th style="text-align:left;"> recall </th>
##    <th style="text-align:left;"> f1-score </th>
##    <th style="text-align:left;"> support </th>
##   </tr>
##  </thead>
## <tbody>
##   <tr>
##    <td style="text-align:left;"> 0 </td>
##    <td style="text-align:left;"> 0.871505 </td>
##    <td style="text-align:left;"> 0.607380 </td>
##    <td style="text-align:left;"> 0.715856 </td>
##    <td style="text-align:left;"> 2412.000000 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> 1 </td>
##    <td style="text-align:left;"> 0.479384 </td>
##    <td style="text-align:left;"> 0.801471 </td>
##    <td style="text-align:left;"> 0.599931 </td>
##    <td style="text-align:left;"> 1088.000000 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> accuracy </td>
##    <td style="text-align:left;"> 0.667714 </td>
##    <td style="text-align:left;"> 0.667714 </td>
##    <td style="text-align:left;"> 0.667714 </td>
##    <td style="text-align:left;"> 3500.000000 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> macro avg </td>
##    <td style="text-align:left;"> 0.675445 </td>
##    <td style="text-align:left;"> 0.704425 </td>
##    <td style="text-align:left;"> 0.657894 </td>
##    <td style="text-align:left;"> 3500.000000 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> weighted avg </td>
##    <td style="text-align:left;"> 0.749612 </td>
##    <td style="text-align:left;"> 0.667714 </td>
##    <td style="text-align:left;"> 0.679820 </td>
##    <td style="text-align:left;"> 3500.000000 </td>
##   </tr>
## </tbody>
## </table>
## 
## Confusion Matrix
## <table>
##  <thead>
##   <tr>
##    <th style="text-align:left;">   </th>
##    <th style="text-align:right;"> Positive (1) </th>
##    <th style="text-align:right;"> Negative (0) </th>
##   </tr>
##  </thead>
## <tbody>
##   <tr>
##    <td style="text-align:left;"> Positive (1) </td>
##    <td style="text-align:right;"> 872 </td>
##    <td style="text-align:right;"> 947 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Negative (0) </td>
##    <td style="text-align:right;"> 216 </td>
##    <td style="text-align:right;"> 1465 </td>
##   </tr>
## </tbody>
## </table>
## 
## AUC/AUPRC
## AUC (ROC): 0.809206
## AUPRC: 0.713748
## 
## ===== VALIDATION =====
## 
## Classification Report
## <table>
##  <thead>
##   <tr>
##    <th style="text-align:left;">   </th>
##    <th style="text-align:left;"> precision </th>
##    <th style="text-align:left;"> recall </th>
##    <th style="text-align:left;"> f1-score </th>
##    <th style="text-align:left;"> support </th>
##   </tr>
##  </thead>
## <tbody>
##   <tr>
##    <td style="text-align:left;"> 0 </td>
##    <td style="text-align:left;"> 0.892537 </td>
##    <td style="text-align:left;"> 0.610204 </td>
##    <td style="text-align:left;"> 0.724848 </td>
##    <td style="text-align:left;"> 490.000000 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> 1 </td>
##    <td style="text-align:left;"> 0.539759 </td>
##    <td style="text-align:left;"> 0.861538 </td>
##    <td style="text-align:left;"> 0.663704 </td>
##    <td style="text-align:left;"> 260.000000 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> accuracy </td>
##    <td style="text-align:left;"> 0.697333 </td>
##    <td style="text-align:left;"> 0.697333 </td>
##    <td style="text-align:left;"> 0.697333 </td>
##    <td style="text-align:left;"> 750.000000 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> macro avg </td>
##    <td style="text-align:left;"> 0.716148 </td>
##    <td style="text-align:left;"> 0.735871 </td>
##    <td style="text-align:left;"> 0.694276 </td>
##    <td style="text-align:left;"> 750.000000 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> weighted avg </td>
##    <td style="text-align:left;"> 0.770241 </td>
##    <td style="text-align:left;"> 0.697333 </td>
##    <td style="text-align:left;"> 0.703652 </td>
##    <td style="text-align:left;"> 750.000000 </td>
##   </tr>
## </tbody>
## </table>
## 
## Confusion Matrix
## <table>
##  <thead>
##   <tr>
##    <th style="text-align:left;">   </th>
##    <th style="text-align:right;"> Positive (1) </th>
##    <th style="text-align:right;"> Negative (0) </th>
##   </tr>
##  </thead>
## <tbody>
##   <tr>
##    <td style="text-align:left;"> Positive (1) </td>
##    <td style="text-align:right;"> 224 </td>
##    <td style="text-align:right;"> 191 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Negative (0) </td>
##    <td style="text-align:right;"> 36 </td>
##    <td style="text-align:right;"> 299 </td>
##   </tr>
## </tbody>
## </table>
## 
## AUC/AUPRC
## AUC (ROC): 0.835290
## AUPRC: 0.766330
## 
## ===== TEST =====
## 
## Classification Report
## <table>
##  <thead>
##   <tr>
##    <th style="text-align:left;">   </th>
##    <th style="text-align:left;"> precision </th>
##    <th style="text-align:left;"> recall </th>
##    <th style="text-align:left;"> f1-score </th>
##    <th style="text-align:left;"> support </th>
##   </tr>
##  </thead>
## <tbody>
##   <tr>
##    <td style="text-align:left;"> 0 </td>
##    <td style="text-align:left;"> 0.859944 </td>
##    <td style="text-align:left;"> 0.579245 </td>
##    <td style="text-align:left;"> 0.692221 </td>
##    <td style="text-align:left;"> 530.000000 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> 1 </td>
##    <td style="text-align:left;"> 0.432570 </td>
##    <td style="text-align:left;"> 0.772727 </td>
##    <td style="text-align:left;"> 0.554649 </td>
##    <td style="text-align:left;"> 220.000000 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> accuracy </td>
##    <td style="text-align:left;"> 0.636000 </td>
##    <td style="text-align:left;"> 0.636000 </td>
##    <td style="text-align:left;"> 0.636000 </td>
##    <td style="text-align:left;"> 750.000000 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> macro avg </td>
##    <td style="text-align:left;"> 0.646257 </td>
##    <td style="text-align:left;"> 0.675986 </td>
##    <td style="text-align:left;"> 0.623435 </td>
##    <td style="text-align:left;"> 750.000000 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> weighted avg </td>
##    <td style="text-align:left;"> 0.734581 </td>
##    <td style="text-align:left;"> 0.636000 </td>
##    <td style="text-align:left;"> 0.651867 </td>
##    <td style="text-align:left;"> 750.000000 </td>
##   </tr>
## </tbody>
## </table>
## 
## Confusion Matrix
## <table>
##  <thead>
##   <tr>
##    <th style="text-align:left;">   </th>
##    <th style="text-align:right;"> Positive (1) </th>
##    <th style="text-align:right;"> Negative (0) </th>
##   </tr>
##  </thead>
## <tbody>
##   <tr>
##    <td style="text-align:left;"> Positive (1) </td>
##    <td style="text-align:right;"> 170 </td>
##    <td style="text-align:right;"> 223 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Negative (0) </td>
##    <td style="text-align:right;"> 50 </td>
##    <td style="text-align:right;"> 307 </td>
##   </tr>
## </tbody>
## </table>
## 
## AUC/AUPRC
## AUC (ROC): 0.781407
## AUPRC: 0.654612

Scenario 2 — Saved plots





















































































































































mirror server hosted at Truenetwork, Russian Federation.