Update README - Run 20251012_060013

README.md

@@ -21,7 +21,7 @@ model-index:
       type: imagenet-1k
     metrics:
     - type: accuracy
-      value: 82.79
+      value: 82.98
 ---
 
 # David: Multi-Scale Crystal Classifier
@@ -51,17 +51,19 @@ as class prototypes with role-weighted similarity computation (Rose Loss).
 ## Performance
 
 ### Best Results
-- **Validation Accuracy**: 82.79%
-- **Best Epoch**: 5
-- **Final Train Accuracy**: 88.18%
+- **Validation Accuracy**: 82.98%
+- **Best Epoch**: 7
+- **Final Train Accuracy**: 89.95%
 
 ### Per-Scale Performance
-- **Scale 256**: 82.79%
-- **Scale 512**: 82.85%
-- **Scale 768**: 82.83%
-- **Scale 1024**: 82.80%
-- **Scale 1280**: 82.65%
-- **Scale 1536**: 82.31%
+- **Scale 256**: 82.98%
+- **Scale 512**: 83.15%
+- **Scale 768**: 83.09%
+- **Scale 1024**: 83.08%
+- **Scale 1280**: 82.99%
+- **Scale 1536**: 82.93%
+- **Scale 1792**: 82.88%
+- **Scale 2048**: 82.67%
 
 
 ## Usage
@@ -184,4 +186,4 @@ Special thanks to Claude (Anthropic) for debugging assistance.
 
 ---
 
-*Generated on 2025-10-12 06:31:25*
+*Generated on 2025-10-12 06:35:55*