Update README - Run 20251012_060013

README.md

@@ -12,7 +12,7 @@ datasets:
 metrics:
 - accuracy
 model-index:
-- name: David-fully_shared-weighted_sum
+- name: David-partial_shared-deep_efficiency
   results:
   - task:
       type: image-classification
@@ -21,7 +21,7 @@ model-index:
       type: imagenet-1k
     metrics:
     - type: accuracy
-      value: 75.38
+      value: 80.79
 ---
 
 # David: Multi-Scale Crystal Classifier
@@ -32,32 +32,31 @@ as class prototypes with role-weighted similarity computation (Rose Loss).
 ## Model Details
 
 ### Architecture
-- **Preset**: small_fast
-- **Sharing Mode**: fully_shared
-- **Fusion Mode**: weighted_sum
-- **Scales**: [256, 512]
-- **Feature Dim**: 512
+- **Preset**: clip_vit_l14_deep
+- **Sharing Mode**: partial_shared
+- **Fusion Mode**: deep_efficiency
+- **Scales**: [256, 512, 768, 1024, 1280, 1536, 1792, 2048, 2304, 2560]
+- **Feature Dim**: 768
 - **Parameters**: ~8.8M
 
 ### Training Configuration
 - **Dataset**: AbstractPhil/imagenet-clip-features-orderly
-- **Model Variant**: clip_vit_b16
+- **Model Variant**: clip_vit_l14
 - **Epochs**: 10
 - **Batch Size**: 1024
-- **Learning Rate**: 0.01
+- **Learning Rate**: 0.001
 - **Rose Loss Weight**: 0.1 → 0.5
 - **Cayley Loss**: False
 
 ## Performance
 
 ### Best Results
-- **Validation Accuracy**: 75.38%
-- **Best Epoch**: 9
-- **Final Train Accuracy**: 76.72%
+- **Validation Accuracy**: 80.79%
+- **Best Epoch**: 0
+- **Final Train Accuracy**: 77.21%
 
 ### Per-Scale Performance
-- **Scale 256**: 75.10%
-- **Scale 512**: 75.30%
+- **Scale 256**: 80.79%
 
 
 ## Usage
@@ -123,7 +122,7 @@ with torch.no_grad():
 ## Architecture Overview
 
 ### Multi-Scale Processing
-David processes inputs at multiple scales (256, 512), 
+David processes inputs at multiple scales (256, 512, 768, 1024, 1280, 1536, 1792, 2048, 2304, 2560), 
 allowing it to capture both coarse and fine-grained features.
 
 ### Crystal Geometry
@@ -141,7 +140,7 @@ score = w_anchor * sim(z, anchor) + w_need * sim(z, need) + ...
 ```
 
 ### Fusion Strategy
-**weighted_sum**: Intelligently combines predictions from multiple scales.
+**deep_efficiency**: Intelligently combines predictions from multiple scales.
 
 ## Training Details
 
@@ -180,4 +179,4 @@ Special thanks to Claude (Anthropic) for debugging assistance.
 
 ---
 
-*Generated on 2025-10-12 06:17:02*
+*Generated on 2025-10-12 06:19:35*