Update beatrix-4simplex-45d_20251007_231008 - Epoch 0 - Acc: 0.0546

README.md

@@ -45,41 +45,42 @@ When train accuracy exceeds validation accuracy by 2% or more, the system automa
 | beatrix-cifar100 | 20251007_195812 | 0.4701 | 42 | `weights/beatrix-cifar100/20251007_195812` | `logs/beatrix-cifar100/20251007_195812` |
 | beatrix-cifar100 | 20251007_203741 | 0.4324 | 40 | `weights/beatrix-cifar100/20251007_203741` | `logs/beatrix-cifar100/20251007_203741` |
 | beatrix-cifar100 | 20251007_193112 | 0.2802 | 10 | `weights/beatrix-cifar100/20251007_193112` | `N/A` |
+| beatrix-4simplex-45d | 20251007_231008 | 0.0546 | 0 | `weights/beatrix-4simplex-45d/20251007_231008` | `logs/beatrix-4simplex-45d/20251007_231008` |
 
 
-## Latest Updated Model: beatrix-cifar100 (Session: 20251007_215344)
+## Latest Updated Model: beatrix-4simplex-45d (Session: 20251007_231008)
 
 ### Model Details
 
 - **Architecture**: Vision Transformer with fractal positional encoding
 - **Dataset**: CIFAR-100 (100 classes)
-- **Embedding Dimension**: 512
-- **Depth**: 12 layers
+- **Embedding Dimension**: 45
+- **Depth**: 22 layers
 - **Patch Size**: 4x4
 - **PE Levels**: 12
-- **Simplex Dimension**: 5-simplex
+- **Simplex Dimension**: 4-simplex
 
 ### Training Details
 
-- **Training Session**: 20251007_215344
-- **Best Accuracy**: 0.5161
-- **Epochs Trained**: 41
+- **Training Session**: 20251007_231008
+- **Best Accuracy**: 0.0546
+- **Epochs Trained**: 0
 - **Batch Size**: 512
 - **Learning Rate**: 0.0001
 - **Adaptive Augmentation**: Enabled
 
 ### Loss Configuration
 
-- Task Loss Weight: 0.5
-- Flow Alignment Weight: 0.5
-- Coherence Weight: 0.3
-- Multi-Scale Weight: 0.2
+- Task Loss Weight: 1.0
+- Flow Alignment Weight: 0.4
+- Coherence Weight: 0.2
+- Multi-Scale Weight: 0.1
 
 ### TensorBoard Logs
 
 Training logs are available in the repository at:
 ```
-logs/beatrix-cifar100/20251007_215344
+logs/beatrix-4simplex-45d/20251007_231008
 ```
 
 To view them locally:
@@ -88,7 +89,7 @@ To view them locally:
 git clone https://huggingface.co/AbstractPhil/vit-beatrix
 
 # View logs in TensorBoard
-tensorboard --logdir vit-beatrix/logs/beatrix-cifar100/20251007_215344
+tensorboard --logdir vit-beatrix/logs/beatrix-4simplex-45d/20251007_231008
 ```
 
 ## Usage
@@ -133,18 +134,18 @@ with open(manifest_path, 'r') as f:
 for key, info in sorted(manifest.items(), key=lambda x: x[1]['accuracy'], reverse=True):
     print(f"{info['model_name']} ({info['timestamp']}): {info['accuracy']:.4f}")
 
-# Download weights for the latest training session of beatrix-cifar100
+# Download weights for the latest training session of beatrix-4simplex-45d
 weights_path = hf_hub_download(
     repo_id="AbstractPhil/vit-beatrix",
-    filename="weights/beatrix-cifar100/20251007_215344/model.safetensors"
+    filename="weights/beatrix-4simplex-45d/20251007_231008/model.safetensors"
 )
 
 # Load model
 model = SimplifiedGeometricClassifier(
     num_classes=100,
     img_size=32,
-    embed_dim=512,
-    depth=12
+    embed_dim=45,
+    depth=22
 )
 
 # Load weights

logs/beatrix-4simplex-45d/20251007_231008/events.out.tfevents.1759878610.f41d735e1a7f.86975.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4db2f03892093ee2439929df8c3573f9bf3e753b283014f3669a0aae516d0cc3
+size 88

manifest.json

@@ -55,5 +55,17 @@
     "depth": 12,
     "batch_size": 512,
     "learning_rate": 0.0001
+  },
+  "beatrix-4simplex-45d_20251007_231008": {
+    "accuracy": 0.054647288657724856,
+    "epoch": 0,
+    "timestamp": "20251007_231008",
+    "model_name": "beatrix-4simplex-45d",
+    "path": "weights/beatrix-4simplex-45d/20251007_231008",
+    "logs_path": "logs/beatrix-4simplex-45d/20251007_231008",
+    "embed_dim": 45,
+    "depth": 22,
+    "batch_size": 512,
+    "learning_rate": 0.0001
   }
 }

weights/beatrix-4simplex-45d/20251007_231008/config.json

@@ -0,0 +1,45 @@
+{
+  "num_classes": 100,
+  "img_size": 32,
+  "patch_size": 4,
+  "embed_dim": 45,
+  "k_simplex": 4,
+  "depth": 22,
+  "num_heads": 5,
+  "mlp_ratio": 4.0,
+  "dropout": 0.0,
+  "pe_levels": 12,
+  "pe_features_per_level": 2,
+  "pe_smooth_tau": 0.25,
+  "simplex_feature_weight": 0.5,
+  "batch_size": 512,
+  "num_epochs": 100,
+  "learning_rate": 0.0001,
+  "weight_decay": 0.005,
+  "warmup_epochs": 0,
+  "task_loss_weight": 1.0,
+  "flow_loss_weight": 0.4,
+  "coherence_loss_weight": 0.2,
+  "multiscale_loss_weight": 0.1,
+  "volume_reg_weight": 0.1,
+  "use_adaptive_augmentation": true,
+  "overfit_threshold": 0.02,
+  "mixup_alpha": 0.2,
+  "cutmix_alpha": 1.0,
+  "device": "cuda",
+  "num_workers": 4,
+  "pin_memory": true,
+  "save_dir": "./checkpoints",
+  "save_every": 10,
+  "use_safetensors": true,
+  "timestamp_dirs": true,
+  "push_to_hub": true,
+  "hub_model_id": "AbstractPhil/vit-beatrix",
+  "hub_model_name": "beatrix-4simplex-45d",
+  "hub_upload_best_only": true,
+  "hub_upload_every_n_epochs": 10,
+  "use_tensorboard": true,
+  "log_dir": "./logs",
+  "log_every": 50,
+  "train_baseline": false
+}

weights/beatrix-4simplex-45d/20251007_231008/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc6cb7b5f728ff408e45ebc5613e9602d070d4776a1a335ecf463c0f1d415f96
+size 2248960