Update MODEL_SUMMARY.txt - Run 20251012_221046

weights/David-decoupled-deep_efficiency/20251012_221046/MODEL_SUMMARY.txt

@@ -3,24 +3,25 @@
 ║                      DAVID MODEL SUMMARY                      ║
 ╠══════════════════════════════════════════════════════════════╣
 ║                                                               ║
-║  🎯 VALIDATION ACCURACY: 64.04%                          ║
+║  🎯 VALIDATION ACCURACY: 64.86%                          ║
 ║                                                               ║
 ╚══════════════════════════════════════════════════════════════╝
 
 MODEL: David-decoupled-deep_efficiency
 RUN ID: 20251012_221046
-BEST EPOCH: 2/10
+BEST EPOCH: 3/10
 
 ═══════════════════════════════════════════════════════════════
 
 📊 PERFORMANCE BREAKDOWN
 
-Final Training Accuracy:  61.67%
-Best Validation Accuracy: 64.04%
+Final Training Accuracy:  62.79%
+Best Validation Accuracy: 64.86%
 
 Per-Scale Accuracies:
-  • Scale  256: 64.04%
-  • Scale  512: 69.15%
+  • Scale  256: 64.86%
+  • Scale  512: 70.63%
+  • Scale  768: 70.71%
 
 ═══════════════════════════════════════════════════════════════
 
@@ -40,17 +41,18 @@ Parameters:    14,877,593
 Epoch | Train Acc | Val Acc  | Learning Rate
 ------|-----------|----------|--------------
     1 |    56.43% |   62.52% | 9.76e-03
-    2 |    61.67% |   64.04% 👑 | 9.05e-03
+    2 |    61.67% |   64.04% | 9.05e-03
+    3 |    62.79% |   64.86% 👑 | 7.94e-03
 
 ═══════════════════════════════════════════════════════════════
 
 📁 FILES
 
-Best Model:    best_model_acc64.04.safetensors
+Best Model:    best_model_acc64.86.safetensors
 Config:        david_config.json
 Training Cfg:  train_config.json
 History:       training_history.json
 
 ═══════════════════════════════════════════════════════════════
 
-Generated: 2025-10-12 22:20:10
+Generated: 2025-10-12 22:24:46