Update README - Run 20251012_221046

README.md

@@ -12,7 +12,7 @@ datasets:
 metrics:
 - accuracy
 model-index:
-- name: David-partial_shared-hierarchical_tree
+- name: David-decoupled-deep_efficiency
   results:
   - task:
       type: image-classification
@@ -21,7 +21,7 @@ model-index:
       type: imagenet-1k
     metrics:
     - type: accuracy
-      value: 74.23
+      value: 62.52
 ---
 
 # David: Multi-Scale Feature Classifier
@@ -36,12 +36,12 @@ exist simultaneously in the same shared space with the correct checks and spacin
 ## Model Details
 
 ### Architecture
-- **Preset**: balanced
-- **Sharing Mode**: partial_shared
-- **Fusion Mode**: hierarchical_tree
-- **Scales**: [256, 512, 768, 1024]
+- **Preset**: high_accuracy
+- **Sharing Mode**: decoupled
+- **Fusion Mode**: deep_efficiency
+- **Scales**: [256, 512, 768, 1024, 1280]
 - **Feature Dim**: 512
-- **Parameters**: 8,758,271
+- **Parameters**: 14,877,593
 
 ### Training Configuration
 - **Dataset**: AbstractPhil/imagenet-clip-features-orderly
@@ -55,14 +55,12 @@ exist simultaneously in the same shared space with the correct checks and spacin
 ## Performance
 
 ### Best Results
-- **Validation Accuracy**: 74.23%
-- **Best Epoch**: 7
-- **Final Train Accuracy**: 80.85%
+- **Validation Accuracy**: 62.52%
+- **Best Epoch**: 0
+- **Final Train Accuracy**: 56.43%
 
 ### Per-Scale Performance
-- **Scale 256**: 71.20%
-- **Scale 512**: 74.21%
-- **Scale 768**: 74.31%
+- **Scale 256**: 62.52%
 
 
 ## Usage
@@ -79,19 +77,19 @@ AbstractPhil/david-shared-space/
 ├── README.md                         # This file
 ├── best_model.json                   # Latest best model info
 ├── weights/
-│   └── david_balanced/
-│       └── 20251012_210041/
+│   └── david_high_accuracy/
+│       └── 20251012_221046/
 │           ├── MODEL_SUMMARY.txt     # 🎯 Human-readable performance summary
 │           ├── training_history.json # 📈 Epoch-by-epoch training curve
-│           ├── best_model_acc74.23.safetensors  # ⭐ Accuracy in filename!
-│           ├── best_model_acc74.23_metadata.json
+│           ├── best_model_acc62.52.safetensors  # ⭐ Accuracy in filename!
+│           ├── best_model_acc62.52_metadata.json
 │           ├── final_model.safetensors
 │           ├── checkpoint_epoch_X_accYY.YY.safetensors
 │           ├── david_config.json
 │           └── train_config.json
 └── runs/
-    └── david_balanced/
-        └── 20251012_210041/
+    └── david_high_accuracy/
+        └── 20251012_221046/
             └── events.out.tfevents.* # TensorBoard logs
 ```
 
@@ -104,9 +102,9 @@ from huggingface_hub import hf_hub_download
 # Browse available models in MODELS_INDEX.json first!
 
 # Specify model variant and run
-model_name = "david_balanced"
-run_id = "20251012_210041"
-accuracy = "74.23"  # From MODELS_INDEX.json
+model_name = "david_high_accuracy"
+run_id = "20251012_221046"
+accuracy = "62.52"  # From MODELS_INDEX.json
 
 # Download config
 config_path = hf_hub_download(
@@ -155,7 +153,7 @@ with torch.no_grad():
 ## Architecture Overview
 
 ### Multi-Scale Processing
-David processes inputs at multiple scales (256, 512, 768, 1024), 
+David processes inputs at multiple scales (256, 512, 768, 1024, 1280), 
 allowing it to capture both coarse and fine-grained features.
 
 ### Shared Representation Space
@@ -176,7 +174,7 @@ score = w_anchor * sim(z, anchor) + w_need * sim(z, need) + ...
 ```
 
 ### Fusion Strategy
-**hierarchical_tree**: Intelligently combines predictions from multiple scales.
+**deep_efficiency**: Intelligently combines predictions from multiple scales.
 
 ## Training Details
 
@@ -200,7 +198,7 @@ score = w_anchor * sim(z, anchor) + w_need * sim(z, need) + ...
   author = {AbstractPhil},
   year = {2025},
   url = {https://huggingface.co/AbstractPhil/david-shared-space},
-  note = {Run ID: 20251012_210041}
+  note = {Run ID: 20251012_221046}
 }
 ```
 
@@ -215,4 +213,4 @@ Special thanks to Claude (Anthropic) for debugging assistance.
 
 ---
 
-*Generated on 2025-10-12 21:45:35*
+*Generated on 2025-10-12 22:15:33*