Update README - Run 20251012_065325

README.md

@@ -21,7 +21,7 @@ model-index:
       type: imagenet-1k
     metrics:
     - type: accuracy
-      value: 71.23
+      value: 71.67
 ---
 
 # David: Multi-Scale Crystal Classifier
@@ -51,12 +51,12 @@ as class prototypes with role-weighted similarity computation (Rose Loss).
 ## Performance
 
 ### Best Results
-- **Validation Accuracy**: 71.23%
-- **Best Epoch**: 1
-- **Final Train Accuracy**: 71.58%
+- **Validation Accuracy**: 71.67%
+- **Best Epoch**: 2
+- **Final Train Accuracy**: 72.76%
 
 ### Per-Scale Performance
-- **Scale 256**: 71.23%
+- **Scale 256**: 71.67%
 
 
 ## Usage
@@ -190,4 +190,4 @@ Special thanks to Claude (Anthropic) for debugging assistance.
 
 ---
 
-*Generated on 2025-10-12 06:59:28*
+*Generated on 2025-10-12 07:01:07*