Epoch 50: 31.90%

README.md

@@ -5,7 +5,7 @@ tags:
 - cifar100
 - geometric-learning
 - fractal-encoding
-- trained
+- in-training
 - no-attention
 - no-cross-entropy
 datasets:
@@ -15,7 +15,7 @@ metrics:
 library_name: pytorch
 pipeline_tag: image-classification
 model-index:
-- name: geo-beatrix-resnet34-step12-feats100
+- name: geo-beatrix-resnet34-step20-feats1000
   results:
   - task:
       type: image-classification
@@ -30,13 +30,13 @@ model-index:
       verified: false
 ---
 
-# geo-beatrix-resnet34-step12-feats100
+# geo-beatrix-resnet34-step20-feats1000
 
 **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** | 57.80% |
-| **Current Test Accuracy** | 42.63% |
-| **Current α (Cantor param)** | 0.4994 |
-| **Total Parameters** | 21,846,001 |
-| **Training Time** | 0:20:47 |
+| **Current Train Accuracy** | 39.59% |
+| **Current Test Accuracy** | 31.90% |
+| **Current α (Cantor param)** | 0.4103 |
+| **Total Parameters** | 28,561,101 |
+| **Training Time** | 0:05:11 |
 
 ### All Training Runs
 
@@ -65,6 +65,7 @@ Final Status: Epoch 200/200
 | `20251011_012453` | ✅ | 160 | **56.12%** | 40.18% | 0.4963 |
 | `20251011_023128` | ✅ | 160 | **56.12%** | 54.65% | 0.5005 |
 | `20251011_025919` | ✅ | 160 | **56.12%** | 57.80% | 0.4994 |
+| `20251011_032343` | 🔄 | 160 | **56.12%** | 39.59% | 0.4103 |
 | `20251010_200842` | ✅ | 180 | **53.61%** | 67.53% | 0.4442 |
 | `20251010_185133` | ✅ | 200 | **52.97%** | 69.87% | 0.4452 |
 
@@ -72,7 +73,7 @@ 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%
@@ -85,8 +86,8 @@ Final Status: Epoch 200/200
 - **Pretrained**: From scratch
 - **Features**: 512-dim from ResNet34
 - **Positional Encoding**: Devil's Staircase (Cantor function, 1883)
-- **PE Levels**: 12
-- **PE Features/Level**: 100
+- **PE Levels**: 20
+- **PE Features/Level**: 1000
 - **Classification**: Geometric Basin Compatibility (NO cross-entropy)
 - **Attention Mechanisms**: NONE
 - **Mixing**: Fractal (triadic multi-patch)
@@ -97,7 +98,7 @@ Final Status: Epoch 200/200
 
 ```json
 {
-  "model_name": "geo-beatrix-resnet34-step12-feats100",
+  "model_name": "geo-beatrix-resnet34-step20-feats1000",
   "model_type": "geometric_basin_classifier",
   "num_classes": 100,
   "batch_size": 512,
@@ -105,8 +106,8 @@ Final Status: Epoch 200/200
   "base_learning_rate": 0.001,
   "weight_decay": 0.05,
   "warmup_epochs": 10,
-  "pe_levels": 12,
-  "pe_features_per_level": 100,
+  "pe_levels": 20,
+  "pe_features_per_level": 1000,
   "dropout": 0.1,
   "pretrained_resnet": false,
   "frozen_resnet": false,
@@ -140,7 +141,7 @@ Final Status: Epoch 200/200
   "loss_function": "Geometric Basin Compatibility",
   "cross_entropy": false,
   "attention_mechanisms": false,
-  "timestamp": "20251011_025919"
+  "timestamp": "20251011_032343"
 }
 ```
 
@@ -154,7 +155,7 @@ 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_025919/
+├── weights/geo-beatrix-resnet34-step20-feats1000/20251011_032343/
 │   ├── model.pt                 (best from this training run)
 │   ├── model.safetensors        (best from this training run)
 │   ├── config.json
@@ -164,7 +165,7 @@ Final Status: Epoch 200/200
 │       ├── checkpoint_epoch_100.safetensors
 │       └── checkpoint_epoch_150.safetensors
 │       (snapshots every 10 epochs)
-└── runs/geo-beatrix-resnet34-step12-feats100/20251011_025919/
+└── runs/geo-beatrix-resnet34-step20-feats1000/20251011_032343/
     ├── events.out.tfevents.*    (TensorBoard logs)
     └── metrics.csv              (training metrics)
 ```
@@ -200,13 +201,13 @@ 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_025919/model.safetensors"
+    filename="weights/geo-beatrix-resnet34-step20-feats1000/20251011_032343/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_025919/checkpoints/checkpoint_epoch_100.safetensors"
+    filename="weights/geo-beatrix-resnet34-step20-feats1000/20251011_032343/checkpoints/checkpoint_epoch_100.safetensors"
 )
 ```
 
@@ -216,23 +217,21 @@ epoch_checkpoint = hf_hub_download(
 
 ### Best Checkpoint
 - Epoch: 160
-- Train Acc: 54.78%
-- Test Acc: 40.25%
-- Alpha: 0.5023
-- Loss: 0.8972
+- Train Acc: 39.59%
+- Test Acc: 56.12%
+- Alpha: 0.4103
+- Loss: 0.0000
 
 ### Latest 5 Epochs
 
-- **Epoch 196**: Train 59.74%, Test 0.00%, α=0.4994, Loss=0.9041
-- **Epoch 197**: Train 57.60%, Test 0.00%, α=0.4991, Loss=0.8370
-- **Epoch 198**: Train 57.25%, Test 0.00%, α=0.4993, Loss=0.8311
-- **Epoch 199**: Train 60.42%, Test 0.00%, α=0.4995, Loss=0.9093
-- **Epoch 200**: Train 57.80%, Test 42.63%, α=0.4994, Loss=0.8336
+- **Epoch 46**: Train 37.85%, Test 0.00%, α=0.4061, Loss=1.8099
+- **Epoch 47**: Train 38.80%, Test 0.00%, α=0.4090, Loss=1.7620
+- **Epoch 48**: Train 38.73%, Test 0.00%, α=0.4112, Loss=1.7839
+- **Epoch 49**: Train 38.73%, Test 0.00%, α=0.4120, Loss=1.7659
+- **Epoch 50**: Train 39.59%, Test 31.90%, α=0.4103, Loss=1.7749
 
 ### Training Milestones
-- 📊 **α ≥ 0.40** reached at epoch 1
-- 📊 **α ≥ 0.44** (near triadic equilibrium) at epoch 1
-- ⚛️  **α = 0.50** (TRIADIC EQUILIBRIUM!) at epoch 2
+- 📊 **α ≥ 0.40** reached at epoch 40
 
 ---