Add training data

training_data.json

@@ -1,37 +1,37 @@
 {
   "train_losses": [
-    158443.7097314453,
-    79525.31055175781,
-    55150.76360839844,
-    27544.720600585937,
-    21215.07262207031,
-    15201.550209960937,
-    17076.67155517578,
-    13131.717595214845,
-    11821.861442871093,
-    11532.774877929687,
-    12098.519575195312,
-    10800.846364746094,
-    11168.50675048828,
-    11143.719829101563,
-    10742.088754882812
+    151209.8428808594,
+    96542.593515625,
+    64421.11040039062,
+    33749.90318359375,
+    19004.055146484374,
+    16831.15723876953,
+    15080.03873046875,
+    12901.305888671875,
+    12439.551579589845,
+    11211.600803222656,
+    12005.076337890625,
+    9869.182932128906,
+    9778.175876464844,
+    9639.344291992187,
+    9386.756044921874
   ],
   "test_losses": [
-    38101.2255859375,
-    32535.043115234374,
-    12535.003198242188,
-    7116.39462890625,
-    8404.210815429688,
-    5529.595751953125,
-    5309.565930175781,
-    5373.602978515625,
-    5507.798205566406,
-    5121.186975097657,
-    4946.643518066407,
-    4945.580090332031,
-    5142.907373046875,
-    5019.152258300781,
-    4867.249340820313
+    58844.00859375,
+    26590.416088867187,
+    6761.0640869140625,
+    5468.9392578125,
+    5257.670593261719,
+    4997.675915527344,
+    5334.108666992188,
+    4735.450335693359,
+    4672.923291015625,
+    4978.906188964844,
+    4423.438391113281,
+    4382.65751953125,
+    4135.9177734375,
+    4164.897058105469,
+    4150.397497558593
   ],
   "config": {
     "epochs": 15,
@@ -44,30 +44,30 @@
       "latent_dim": 96,
       "encoder_dropout": 0.1,
       "decoder_dropout": 0.1,
-      "initial_mi_beta": 0.0,
-      "final_mi_beta": 0.0,
+      "initial_mi_beta": 1.0,
+      "final_mi_beta": 1.0,
       "mi_beta_shape": "constant",
-      "initial_tc_beta": 10.0,
-      "final_tc_beta": 10.0,
+      "initial_tc_beta": 5.0,
+      "final_tc_beta": 5.0,
       "tc_beta_shape": "constant",
       "initial_dw_beta": 0.2,
       "final_dw_beta": 1.0,
-      "dw_beta_shape": "linear",
+      "dw_beta_shape": "custom",
       "warmup_epoch_ratio": 0.2,
       "free_bits": 0.15,
       "focal_loss_alpha": 0.5,
       "focal_loss_gamma": 2.0
     },
     "adaptive_weighting": {
-      "initial_mi_beta": 0.0,
-      "final_mi_beta": 0.0,
+      "initial_mi_beta": 1.0,
+      "final_mi_beta": 1.0,
       "mi_beta_shape": "constant",
-      "initial_tc_beta": 10.0,
-      "final_tc_beta": 10.0,
+      "initial_tc_beta": 5.0,
+      "final_tc_beta": 5.0,
       "tc_beta_shape": "constant",
       "initial_dw_beta": 0.2,
       "final_dw_beta": 1.0,
-      "dw_beta_shape": "linear",
+      "dw_beta_shape": "custom",
       "warmup_epoch_ratio": 0.2
     },
     "regularization": {
@@ -85,9 +85,9 @@
       "best_epoch": null
     }
   },
-  "final_train_loss": 10742.088754882812,
-  "final_test_loss": 4867.249340820313,
+  "final_train_loss": 9386.756044921874,
+  "final_test_loss": 4150.397497558593,
   "total_epochs": 15,
-  "best_train_loss": 10742.088754882812,
-  "best_test_loss": 4867.249340820313
+  "best_train_loss": 9386.756044921874,
+  "best_test_loss": 4135.9177734375
 }