Add training data

training_data.json

@@ -1,37 +1,13 @@
 {
   "train_losses": [
-    8522.687722167968,
-    5948.037609863281,
-    5299.650319824219,
-    5164.902270507812,
-    5124.975241699219,
-    4882.509072265625,
-    4695.139870605469,
-    4494.95626953125,
-    3769.9364306640623,
-    3455.5234362792967,
-    3113.4037646484376,
-    2894.6963244628905,
-    2755.5655737304687,
-    2669.2157153320313,
-    2610.3932897949217
+    8604.43171875,
+    5975.3112109375,
+    5188.482888183594
   ],
   "test_losses": [
-    5174.157824707031,
-    4697.121484375,
-    4170.104797363281,
-    4119.477563476563,
-    4505.739526367188,
-    4108.995947265625,
-    4981.743469238281,
-    4981.943542480469,
-    5563.02421875,
-    6102.1019287109375,
-    6386.164367675781,
-    6654.9200439453125,
-    7023.874938964844,
-    7218.628796386719,
-    7328.070239257812
+    5092.959655761719,
+    4591.594555664063,
+    4268.6180419921875
   ],
   "config": {
     "epochs": 15,
@@ -50,7 +26,7 @@
       "initial_dw_beta": 0.02,
       "final_dw_beta": 1.0,
       "dw_beta_shape": "linear",
-      "warmup_epoch_ratio": 0.4
+      "warmup_epoch_ratio": 0.2
     },
     "free_bits": 0.15,
     "focal_loss_alpha": 0.75,
@@ -59,16 +35,16 @@
     "enable_dropout_on_latent": true,
     "enable_dropout_on_decoder": true,
     "early_stopping": {
-      "enabled": false,
+      "enabled": true,
       "patience": 3,
       "min_delta": 0.01,
-      "triggered": false,
-      "best_epoch": null
+      "triggered": true,
+      "best_epoch": 3
     }
   },
-  "final_train_loss": 2610.3932897949217,
-  "final_test_loss": 7328.070239257812,
-  "total_epochs": 15,
-  "best_train_loss": 2610.3932897949217,
-  "best_test_loss": 4108.995947265625
+  "final_train_loss": 5188.482888183594,
+  "final_test_loss": 4268.6180419921875,
+  "total_epochs": 3,
+  "best_train_loss": 5188.482888183594,
+  "best_test_loss": 4268.6180419921875
 }