Update beatrix-simplex7-patch4-256d_20251008_020048 - Epoch 0 - Acc: 0.0552

README.md

@@ -60,15 +60,16 @@ When train accuracy exceeds validation accuracy by 2% or more, the system automa
 | beatrix-4simplex-45d | 20251007_231008 | 0.2916 | 85 | `weights/beatrix-4simplex-45d/20251007_231008` | `logs/beatrix-4simplex-45d/20251007_231008` |
 | beatrix-cifar100 | 20251007_193112 | 0.2802 | 10 | `weights/beatrix-cifar100/20251007_193112` | `N/A` |
 | beatrix-4simplex-45d | 20251008_001147 | 0.1382 | 10 | `weights/beatrix-4simplex-45d/20251008_001147` | `logs/beatrix-4simplex-45d/20251008_001147` |
+| beatrix-simplex7-patch4-256d | 20251008_020048 | 0.0552 | 0 | `weights/beatrix-simplex7-patch4-256d/20251008_020048` | `logs/beatrix-simplex7-patch4-256d/20251008_020048` |
 
 
-## Latest Updated Model: beatrix-simplex7-patch4-45d (Session: 20251008_010524)
+## Latest Updated Model: beatrix-simplex7-patch4-256d (Session: 20251008_020048)
 
 ### Model Details
 
 - **Architecture**: Vision Transformer with fractal positional encoding
 - **Dataset**: CIFAR-100 (100 classes)
-- **Embedding Dimension**: 45
+- **Embedding Dimension**: 256
 - **Depth**: 22 layers
 - **Patch Size**: 4x4
 - **PE Levels**: 12
@@ -77,9 +78,9 @@ When train accuracy exceeds validation accuracy by 2% or more, the system automa
 
 ### Training Details
 
-- **Training Session**: 20251008_010524
-- **Best Accuracy**: 0.2917
-- **Epochs Trained**: 95
+- **Training Session**: 20251008_020048
+- **Best Accuracy**: 0.0552
+- **Epochs Trained**: 0
 - **Batch Size**: 512
 - **Learning Rate**: 0.0001
 - **Adaptive Augmentation**: Enabled
@@ -95,7 +96,7 @@ When train accuracy exceeds validation accuracy by 2% or more, the system automa
 
 Training logs are available in the repository at:
 ```
-logs/beatrix-simplex7-patch4-45d/20251008_010524
+logs/beatrix-simplex7-patch4-256d/20251008_020048
 ```
 
 To view them locally:
@@ -104,7 +105,7 @@ To view them locally:
 git clone https://huggingface.co/AbstractPhil/vit-beatrix
 
 # View logs in TensorBoard
-tensorboard --logdir vit-beatrix/logs/beatrix-simplex7-patch4-45d/20251008_010524
+tensorboard --logdir vit-beatrix/logs/beatrix-simplex7-patch4-256d/20251008_020048
 ```
 
 ## Usage
@@ -149,17 +150,17 @@ 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-simplex7-patch4-45d
+# Download weights for the latest training session of beatrix-simplex7-patch4-256d
 weights_path = hf_hub_download(
     repo_id="AbstractPhil/vit-beatrix",
-    filename="weights/beatrix-simplex7-patch4-45d/20251008_010524/model.safetensors"
+    filename="weights/beatrix-simplex7-patch4-256d/20251008_020048/model.safetensors"
 )
 
 # Load model
 model = SimplifiedGeometricClassifier(
     num_classes=100,
     img_size=32,
-    embed_dim=45,
+    embed_dim=256,
     depth=22
 )
 

logs/beatrix-simplex7-patch4-256d/20251008_020048/events.out.tfevents.1759888850.f41d735e1a7f.160429.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:460e080452cf46e49b5b612e19fdc00264a6fc84408e8ccf1036f958a478df27
+size 88

manifest.json

@@ -103,5 +103,17 @@
     "depth": 22,
     "batch_size": 512,
     "learning_rate": 0.0001
+  },
+  "beatrix-simplex7-patch4-256d_20251008_020048": {
+    "accuracy": 0.05523322615772486,
+    "epoch": 0,
+    "timestamp": "20251008_020048",
+    "model_name": "beatrix-simplex7-patch4-256d",
+    "path": "weights/beatrix-simplex7-patch4-256d/20251008_020048",
+    "logs_path": "logs/beatrix-simplex7-patch4-256d/20251008_020048",
+    "embed_dim": 256,
+    "depth": 22,
+    "batch_size": 512,
+    "learning_rate": 0.0001
   }
 }

weights/beatrix-simplex7-patch4-256d/20251008_020048/config.json

@@ -0,0 +1,47 @@
+{
+  "num_classes": 100,
+  "img_size": 32,
+  "patch_size": 4,
+  "embed_dim": 256,
+  "k_simplex": 7,
+  "depth": 22,
+  "num_heads": 8,
+  "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,
+  "simplex_init_method": "regular",
+  "simplex_init_scale": 1.0,
+  "batch_size": 512,
+  "num_epochs": 100,
+  "learning_rate": 0.0001,
+  "weight_decay": 0.005,
+  "warmup_epochs": 10,
+  "task_loss_weight": 0.5,
+  "flow_loss_weight": 0.5,
+  "coherence_loss_weight": 0.3,
+  "multiscale_loss_weight": 0.2,
+  "volume_reg_weight": 0.1,
+  "use_adaptive_augmentation": true,
+  "overfit_threshold": 0.015,
+  "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-simplex7-patch4-256d",
+  "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-simplex7-patch4-256d/20251008_020048/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af53b96503029fd441e417ec44453dd89dfa7e31716a29f75d48a7d5857d613a
+size 69720924