Epoch 150: 68.72%

README.md

@@ -25,7 +25,7 @@ model-index:
       type: cifar100
     metrics:
     - type: accuracy
-      value: 68.37
+      value: 68.72
       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** | **68.37%** |
-| **Best Epoch** | 195 |
-| **Current Train Accuracy** | 78.33% |
-| **Current Test Accuracy** | 65.48% |
-| **Current α (Cantor param)** | 0.4407 |
+| **Best Test Accuracy** | **68.72%** |
+| **Best Epoch** | 150 |
+| **Current Train Accuracy** | 73.29% |
+| **Current Test Accuracy** | 68.72% |
+| **Current α (Cantor param)** | 0.4477 |
 | **Total Parameters** | 60,440,937 |
-| **Training Time** | 0:31:08 |
+| **Training Time** | 0:46:51 |
 
 ### All Training Runs
 
 | Timestamp | Status | Best Epoch | Test Acc | Train Acc | α |
 |-----------|--------|------------|----------|-----------|---|
-| `20251010_035422` | 🔄 | 195 | **68.37%** | 78.33% | 0.4407 |
+| `20251010_035422` | 🔄 | 150 | **68.72%** | 73.29% | 0.4477 |
 
 ### Comparison to State-of-the-Art
 
 | Model | Accuracy | Status |
 |-------|----------|--------|
-| **geo-beatrix (this model)** | **68.37%** | 🔄 Training |
+| **geo-beatrix (this model)** | **68.72%** | 🔄 Training |
 | vit-beatrix-dualstream | 66.0% | Vision Transformer + Cross-Entropy |
 
 ✅ **geo-beatrix has surpassed all baselines!**
@@ -186,23 +186,25 @@ epoch_checkpoint = hf_hub_download(
 ## Training History
 
 ### Best Checkpoint
-- Epoch: 195
-- Train Acc: 78.33%
-- Test Acc: 68.37%
-- Alpha: 0.4407
-- Loss: 0.0000
+- Epoch: 150
+- Train Acc: 73.29%
+- Test Acc: 68.72%
+- Alpha: 0.4477
+- Loss: 0.5045
 
 ### Latest 5 Epochs
 
-- **Epoch 96**: Train 76.76%, Test 0.00%, α=0.4402, Loss=0.6993
-- **Epoch 97**: Train 79.36%, Test 0.00%, α=0.4385, Loss=0.7110
-- **Epoch 98**: Train 77.46%, Test 0.00%, α=0.4415, Loss=0.6887
-- **Epoch 99**: Train 77.60%, Test 0.00%, α=0.4380, Loss=0.6902
-- **Epoch 100**: Train 78.33%, Test 65.48%, α=0.4407, Loss=0.6848
+- **Epoch 146**: Train 72.97%, Test 0.00%, α=0.4482, Loss=0.5056
+- **Epoch 147**: Train 71.37%, Test 0.00%, α=0.4484, Loss=0.4993
+- **Epoch 148**: Train 74.14%, Test 0.00%, α=0.4469, Loss=0.5086
+- **Epoch 149**: Train 72.99%, Test 0.00%, α=0.4485, Loss=0.5048
+- **Epoch 150**: Train 73.29%, Test 68.72%, α=0.4477, Loss=0.5045
 
 ### Training Milestones
 - 🎯 **50% Accuracy** reached at epoch 35
 - 🎯 **60% Accuracy** reached at epoch 60
+- 🏆 **Beat vit-beatrix (66.0%)** at epoch 110
+- 🚀 **67% Accuracy** reached at epoch 125
 - 📊 **α ≥ 0.40** reached at epoch 9
 - 📊 **α ≥ 0.44** (near triadic equilibrium) at epoch 83