Epoch 150: 66.55%

README.md

@@ -25,7 +25,7 @@ model-index:
       type: cifar100
     metrics:
     - type: accuracy
-      value: 66.72
+      value: 66.74
       name: Test Accuracy
       verified: false
 ---
@@ -36,7 +36,7 @@ model-index:
 
 🚧 **Training in Progress** 🚧
 
-Current Status: Epoch 100/200
+Current Status: Epoch 150/200
 
 ---
 
@@ -44,25 +44,25 @@ Current Status: Epoch 100/200
 
 | Metric | Value |
 |--------|-------|
-| **Best Test Accuracy** | **66.72%** |
-| **Best Epoch** | 190 |
-| **Current Train Accuracy** | 70.49% |
-| **Current Test Accuracy** | 65.23% |
-| **Current α (Cantor param)** | 0.4358 |
+| **Best Test Accuracy** | **66.74%** |
+| **Best Epoch** | 145 |
+| **Current Train Accuracy** | 62.74% |
+| **Current Test Accuracy** | 66.55% |
+| **Current α (Cantor param)** | 0.4408 |
 | **Total Parameters** | 45,235,067 |
-| **Training Time** | 0:27:06 |
+| **Training Time** | 0:40:54 |
 
 ### All Training Runs
 
 | Timestamp | Status | Best Epoch | Test Acc | Train Acc | α |
 |-----------|--------|------------|----------|-----------|---|
-| `20251010_014649` | 🔄 | 190 | **66.72%** | 70.49% | 0.4358 |
+| `20251010_014649` | 🔄 | 145 | **66.74%** | 62.74% | 0.4408 |
 
 ### Comparison to State-of-the-Art
 
 | Model | Accuracy | Status |
 |-------|----------|--------|
-| **geo-beatrix (this model)** | **66.72%** | 🔄 Training |
+| **geo-beatrix (this model)** | **66.74%** | 🔄 Training |
 | vit-beatrix-dualstream | 66.0% | Vision Transformer + Cross-Entropy |
 | CLIP ViT-L/14 (zero-shot) | ~63-65% | 400M image-text pairs |
 | CLIP ViT-B/32 (zero-shot) | ~63.5% | Vision Transformer |
@@ -188,24 +188,26 @@ epoch_checkpoint = hf_hub_download(
 ## Training History
 
 ### Best Checkpoint
-- Epoch: 190
-- Train Acc: 70.49%
-- Test Acc: 66.72%
-- Alpha: 0.4358
-- Loss: 0.0000
+- Epoch: 145
+- Train Acc: 68.64%
+- Test Acc: 66.74%
+- Alpha: 0.4421
+- Loss: 0.5569
 
 ### Latest 5 Epochs
 
-- **Epoch 96**: Train 74.75%, Test 0.00%, α=0.4361, Loss=0.7374
-- **Epoch 97**: Train 72.03%, Test 0.00%, α=0.4375, Loss=0.7204
-- **Epoch 98**: Train 74.35%, Test 0.00%, α=0.4357, Loss=0.7222
-- **Epoch 99**: Train 75.03%, Test 0.00%, α=0.4372, Loss=0.7364
-- **Epoch 100**: Train 70.49%, Test 65.23%, α=0.4358, Loss=0.7120
+- **Epoch 146**: Train 69.81%, Test 0.00%, α=0.4431, Loss=0.5596
+- **Epoch 147**: Train 63.19%, Test 0.00%, α=0.4411, Loss=0.5365
+- **Epoch 148**: Train 67.51%, Test 0.00%, α=0.4416, Loss=0.5459
+- **Epoch 149**: Train 63.99%, Test 0.00%, α=0.4425, Loss=0.5447
+- **Epoch 150**: Train 62.74%, Test 66.55%, α=0.4408, Loss=0.5356
 
 ### Training Milestones
 - 🎯 **50% Accuracy** reached at epoch 40
 - 🎯 **60% Accuracy** reached at epoch 60
+- 🏆 **Beat vit-beatrix (66.0%)** at epoch 145
 - 📊 **α ≥ 0.40** reached at epoch 9
+- 📊 **α ≥ 0.44** (near triadic equilibrium) at epoch 117
 
 ---