End of training

training_summary.json

@@ -2,26 +2,26 @@
   "model_name": "bert-philosophy-classifier",
   "base_model": "maximuspowers/bert-philosophy-adapted",
   "dataset": "maximuspowers/philosophai-papers-labeled",
-  "training_samples": 1159,
-  "validation_samples": 145,
-  "test_samples": 145,
+  "training_samples": 2097,
+  "validation_samples": 262,
+  "test_samples": 263,
   "num_epochs": 500,
   "learning_rate": 2e-05,
   "batch_size": 16,
   "contrastive_weight": 0.2,
   "test_results": {
-    "loss": 0.5930218696594238,
-    "exact_match_accuracy": 0.41379310344827586,
-    "macro_precision": 0.3909600203717851,
-    "macro_recall": 0.2934883988593418,
-    "macro_f1": 0.32054514113337645,
-    "micro_precision": 0.8062015503875969,
-    "micro_recall": 0.5148514851485149,
-    "micro_f1": 0.6283987915407855,
-    "hamming_loss": 0.04989858012170385,
-    "runtime": 0.7302,
-    "samples_per_second": 198.571,
-    "steps_per_second": 26.02
+    "loss": 0.5529536008834839,
+    "exact_match_accuracy": 0.39543726235741444,
+    "macro_precision": 0.48673324944951246,
+    "macro_recall": 0.2821097027243954,
+    "macro_f1": 0.3389521243273359,
+    "micro_precision": 0.8050847457627118,
+    "micro_recall": 0.4947916666666667,
+    "micro_f1": 0.6129032258064516,
+    "hamming_loss": 0.05367926638335943,
+    "runtime": 1.31,
+    "samples_per_second": 200.761,
+    "steps_per_second": 25.191
   },
   "philosophy_schools": [
     "Effective Altruism",