Epoch 150: 51.84%

README.md

@@ -25,7 +25,7 @@ model-index:
       type: cifar100
     metrics:
     - type: accuracy
-      value: 50.42
+      value: 51.91
       name: Test Accuracy
       verified: false
 ---
@@ -36,7 +36,7 @@ model-index:
 
 🚧 **Training in Progress** 🚧
 
-Current Status: Epoch 100/200
+Current Status: Epoch 150/200
 
 ---
 
@@ -44,28 +44,28 @@ Current Status: Epoch 100/200
 
 | Metric | Value |
 |--------|-------|
-| **Best Test Accuracy** | **50.42%** |
-| **Best Epoch** | 95 |
-| **Current Train Accuracy** | 58.72% |
-| **Current Test Accuracy** | 49.53% |
-| **Current α (Cantor param)** | 0.4398 |
+| **Best Test Accuracy** | **51.91%** |
+| **Best Epoch** | 140 |
+| **Current Train Accuracy** | 64.34% |
+| **Current Test Accuracy** | 51.84% |
+| **Current α (Cantor param)** | 0.4450 |
 | **Total Parameters** | 11,952,641 |
-| **Training Time** | 0:15:16 |
+| **Training Time** | 0:23:05 |
 
 ### All Training Runs
 
 | Timestamp | Status | Best Epoch | Test Acc | Train Acc | α |
 |-----------|--------|------------|----------|-----------|---|
-| `20251010_185133` | 🔄 | 95 | **50.42%** | 58.72% | 0.4398 |
+| `20251010_185133` | 🔄 | 140 | **51.91%** | 64.34% | 0.4450 |
 
 ### Comparison to State-of-the-Art
 
 | Model | Accuracy | Status |
 |-------|----------|--------|
-| **geo-beatrix (this model)** | **50.42%** | 🔄 Training |
+| **geo-beatrix (this model)** | **51.91%** | 🔄 Training |
 | vit-beatrix-dualstream | 66.0% | Vision Transformer + Cross-Entropy |
 
-🎯 **Current target**: Beat vit-beatrix (66.0%) - Currently -15.58%
+🎯 **Current target**: Beat vit-beatrix (66.0%) - Currently -14.09%
 
 ---
 
@@ -204,19 +204,19 @@ epoch_checkpoint = hf_hub_download(
 ## Training History
 
 ### Best Checkpoint
-- Epoch: 95
-- Train Acc: 56.55%
-- Test Acc: 50.42%
-- Alpha: 0.4372
-- Loss: 1.1023
+- Epoch: 140
+- Train Acc: 65.59%
+- Test Acc: 51.91%
+- Alpha: 0.4444
+- Loss: 0.8350
 
 ### Latest 5 Epochs
 
-- **Epoch 96**: Train 56.31%, Test 0.00%, α=0.4392, Loss=1.0587
-- **Epoch 97**: Train 58.66%, Test 0.00%, α=0.4378, Loss=1.1022
-- **Epoch 98**: Train 57.02%, Test 0.00%, α=0.4423, Loss=1.0252
-- **Epoch 99**: Train 58.09%, Test 0.00%, α=0.4407, Loss=1.0468
-- **Epoch 100**: Train 58.72%, Test 49.53%, α=0.4398, Loss=1.0677
+- **Epoch 146**: Train 66.88%, Test 0.00%, α=0.4448, Loss=0.8195
+- **Epoch 147**: Train 67.98%, Test 0.00%, α=0.4451, Loss=0.8429
+- **Epoch 148**: Train 68.48%, Test 0.00%, α=0.4468, Loss=0.7858
+- **Epoch 149**: Train 66.43%, Test 0.00%, α=0.4455, Loss=0.7942
+- **Epoch 150**: Train 64.34%, Test 51.84%, α=0.4450, Loss=0.7643
 
 ### Training Milestones
 - 🎯 **50% Accuracy** reached at epoch 95