Add Baseline vs SE validation comparison

evaluation/baseline_vs_se_comparison.json

@@ -43,23 +43,23 @@
             "best_model_epoch": 48,
             "training_reported_best_validation_dice": 0.773046,
             "foreground_macro": {
-                "Dice": 0.751161122192792,
-                "IoU": 0.610363120023954,
-                "Precision": 0.7985494716544408,
-                "Recall": 0.7203059913692811
+                "Dice": 0.7610692589733999,
+                "IoU": 0.6221029962007719,
+                "Precision": 0.8114060556626961,
+                "Recall": 0.7258300177342824
             },
             "tumor": {
-                "Dice": 0.9221925209405566,
-                "IoU": 0.8556189661490515,
-                "Precision": 0.9457929635760306,
-                "Recall": 0.8997412111319005,
-                "Sensitivity": 0.8997412111319005,
-                "FNR": 0.10025878886809707,
-                "FPR": 0.014009904452834464,
-                "TP": 3728802,
-                "TN": 15040639,
-                "FP": 213712,
-                "FN": 415503
+                "Dice": 0.9216496372309628,
+                "IoU": 0.8546847750524318,
+                "Precision": 0.9470200761359134,
+                "Recall": 0.897603068027069,
+                "Sensitivity": 0.897603068027069,
+                "FNR": 0.1023969319729286,
+                "FPR": 0.013822529972448414,
+                "TP": 3758431,
+                "TN": 15001209,
+                "FP": 210261,
+                "FN": 428755
             }
         }
     },