Epoch 50: 33.71%

README.md

@@ -5,7 +5,7 @@ tags:
 - cifar100
 - geometric-learning
 - fractal-encoding
-- trained
+- in-training
 - no-attention
 - no-cross-entropy
 datasets:
@@ -34,9 +34,9 @@ model-index:
 
 **Geometric Basin Classification for CIFAR-100**
 
-🎉 **Training Complete** 🎉
+🚧 **Training in Progress** 🚧
 
-Final Status: Epoch 200/200
+Current Status: Epoch 50/200
 
 ---
 
@@ -46,11 +46,11 @@ Final Status: Epoch 200/200
 |--------|-------|
 | **Best Test Accuracy** | **56.12%** |
 | **Best Epoch** | 160 |
-| **Current Train Accuracy** | 40.18% |
-| **Current Test Accuracy** | 47.65% |
-| **Current α (Cantor param)** | 0.4963 |
+| **Current Train Accuracy** | 41.85% |
+| **Current Test Accuracy** | 33.71% |
+| **Current α (Cantor param)** | 0.4915 |
 | **Total Parameters** | 21,846,001 |
-| **Training Time** | 0:22:58 |
+| **Training Time** | 0:05:35 |
 
 ### All Training Runs
 
@@ -63,6 +63,7 @@ Final Status: Epoch 200/200
 | `20251010_234239` | ✅ | 160 | **56.12%** | 73.34% | 0.4882 |
 | `20251011_002858` | ✅ | 160 | **56.12%** | 46.05% | 0.4712 |
 | `20251011_012453` | ✅ | 160 | **56.12%** | 40.18% | 0.4963 |
+| `20251011_023128` | 🔄 | 160 | **56.12%** | 41.85% | 0.4915 |
 | `20251010_200842` | ✅ | 180 | **53.61%** | 67.53% | 0.4442 |
 | `20251010_185133` | ✅ | 200 | **52.97%** | 69.87% | 0.4452 |
 
@@ -70,7 +71,7 @@ Final Status: Epoch 200/200
 
 | Model | Accuracy | Status |
 |-------|----------|--------|
-| **geo-beatrix (this model)** | **56.12%** | ✅ Complete |
+| **geo-beatrix (this model)** | **56.12%** | 🔄 Training |
 | geo-beatrix (50M params) | 69.0% | Geometric Basin CONV architecture |
 
 🎯 **Current target**: Beat geo-beatrix (69.0%) - Currently -12.88%
@@ -98,10 +99,10 @@ Final Status: Epoch 200/200
   "model_name": "geo-beatrix-resnet34-step12-feats100",
   "model_type": "geometric_basin_classifier",
   "num_classes": 100,
-  "batch_size": 1024,
+  "batch_size": 512,
   "num_epochs": 200,
   "base_learning_rate": 0.001,
-  "weight_decay": 1e-05,
+  "weight_decay": 0.05,
   "warmup_epochs": 10,
   "pe_levels": 12,
   "pe_features_per_level": 100,
@@ -138,7 +139,7 @@ Final Status: Epoch 200/200
   "loss_function": "Geometric Basin Compatibility",
   "cross_entropy": false,
   "attention_mechanisms": false,
-  "timestamp": "20251011_012453"
+  "timestamp": "20251011_023128"
 }
 ```
 
@@ -152,7 +153,7 @@ Final Status: Epoch 200/200
 ├── best_model_info.json     (which epoch/run this came from)
 ├── runs_history.json        (all training runs and their results)
 ├── README.md
-├── weights/geo-beatrix-resnet34-step12-feats100/20251011_012453/
+├── weights/geo-beatrix-resnet34-step12-feats100/20251011_023128/
 │   ├── model.pt                 (best from this training run)
 │   ├── model.safetensors        (best from this training run)
 │   ├── config.json
@@ -162,7 +163,7 @@ Final Status: Epoch 200/200
 │       ├── checkpoint_epoch_100.safetensors
 │       └── checkpoint_epoch_150.safetensors
 │       (snapshots every 10 epochs)
-└── runs/geo-beatrix-resnet34-step12-feats100/20251011_012453/
+└── runs/geo-beatrix-resnet34-step12-feats100/20251011_023128/
     ├── events.out.tfevents.*    (TensorBoard logs)
     └── metrics.csv              (training metrics)
 ```
@@ -198,13 +199,13 @@ with open(info_path) as f:
 # Or download from specific training run
 model_path = hf_hub_download(
     repo_id="AbstractPhil/geo-beatrix-resnet",
-    filename="weights/geo-beatrix-resnet34-step12-feats100/20251011_012453/model.safetensors"
+    filename="weights/geo-beatrix-resnet34-step12-feats100/20251011_023128/model.safetensors"
 )
 
 # Download specific epoch checkpoint
 epoch_checkpoint = hf_hub_download(
     repo_id="AbstractPhil/geo-beatrix-resnet",
-    filename="weights/geo-beatrix-resnet34-step12-feats100/20251011_012453/checkpoints/checkpoint_epoch_100.safetensors"
+    filename="weights/geo-beatrix-resnet34-step12-feats100/20251011_023128/checkpoints/checkpoint_epoch_100.safetensors"
 )
 ```
 
@@ -214,22 +215,22 @@ epoch_checkpoint = hf_hub_download(
 
 ### Best Checkpoint
 - Epoch: 160
-- Train Acc: 40.73%
-- Test Acc: 48.12%
-- Alpha: 0.4966
-- Loss: 1.1347
+- Train Acc: 41.85%
+- Test Acc: 56.12%
+- Alpha: 0.4915
+- Loss: 0.0000
 
 ### Latest 5 Epochs
 
-- **Epoch 196**: Train 42.68%, Test 0.00%, α=0.4963, Loss=1.1746
-- **Epoch 197**: Train 44.84%, Test 0.00%, α=0.4963, Loss=1.3002
-- **Epoch 198**: Train 39.74%, Test 0.00%, α=0.4963, Loss=1.0055
-- **Epoch 199**: Train 37.44%, Test 0.00%, α=0.4963, Loss=0.9007
-- **Epoch 200**: Train 40.18%, Test 47.65%, α=0.4963, Loss=0.9759
+- **Epoch 46**: Train 39.91%, Test 0.00%, α=0.4895, Loss=1.7927
+- **Epoch 47**: Train 40.17%, Test 0.00%, α=0.4897, Loss=1.7402
+- **Epoch 48**: Train 39.89%, Test 0.00%, α=0.4902, Loss=1.6635
+- **Epoch 49**: Train 40.47%, Test 0.00%, α=0.4912, Loss=1.6640
+- **Epoch 50**: Train 41.85%, Test 33.71%, α=0.4915, Loss=1.7345
 
 ### Training Milestones
-- 📊 **α ≥ 0.40** reached at epoch 23
-- 📊 **α ≥ 0.44** (near triadic equilibrium) at epoch 33
+- 📊 **α ≥ 0.40** reached at epoch 14
+- 📊 **α ≥ 0.44** (near triadic equilibrium) at epoch 19
 
 ---