Upload weights, notebooks, sample images

configs/end2end02.yaml

@@ -0,0 +1,280 @@
+### BASELINE: CONVERGES AFTER LONG
+
+parameters:
+
+  ### MODEL ARCHITECTURE                                                       
+  MODEL:
+    value:
+      MODEL_CLASS: "UnReflect_Model_TokenInpainter"  # Main model class name (must match class in models.py)
+      MODEL_MODULE: "models"  # Module name to import model classes from (default: "models")
+      RGB_ENCODER:
+        ENCODER: "facebook/dinov3-vitl16-pretrain-lvd1689m"  # DINOv3 encoder model name (HuggingFace format)
+        IMAGE_SIZE: 896  # Input image size (height and width in pixels)
+        RETURN_SELECTED_LAYERS: [3, 6, 9 , 12]  # Transformer layer indices to extract features from (0-indexed)
+        RGB_ENCODER_LR: 0.0  # Learning rate for RGB encoder (0.0 = frozen, must be explicitly set)
+      DECODERS:
+        diffuse:
+          FEATURE_DIM: 1024  # Feature dimension for decoder (should match encoder output)
+          REASSEMBLE_OUT_CHANNELS: [768,1024,1536,2048]  # Output channels for each decoder stage (DPT-style reassembly)
+          REASSEMBLE_FACTORS: [4.0, 2.0, 1.0, 0.5]  # Spatial upsampling factors for each stage
+          READOUT_TYPE: "ignore"  # Readout type for DPT decoder ("ignore", "project", etc.)
+          FROM_PRETRAINED: "diffuse_decoder.pt"  # Path to pretrained decoder weights (optional)
+          USE_BN: False  # Use batch normalization in decoder
+          DROPOUT: 0.1  # Dropout rate in decoder layers
+          OUTPUT_IMAGE_SIZE: [896,896]  # Output image resolution [height, width]
+          OUTPUT_CHANNELS: 3  # Number of output channels (3 for RGB diffuse image)
+          DECODER_LR: 1.0e-5 # Custom learning rate for decoder (0.0 = frozen, 1.0 = same as base LR)
+          NUM_FUSION_BLOCKS_TRAINABLE: 1  # Number of fusion blocks to train (0-4, null = train all if DECODER_LR != 0)
+          TRAIN_RGB_HEAD: True  # Whether to train RGB head (true/false, null = train if DECODER_LR != 0)
+        highlight:
+          FEATURE_DIM: 1024  # Feature dimension for highlight decoder
+          REASSEMBLE_OUT_CHANNELS: [96,192,384,768]  # Output channels for each decoder stage
+          REASSEMBLE_FACTORS: [4.0, 2.0, 1.0, 0.5]  # Spatial upsampling factors for each stage
+          READOUT_TYPE: "ignore"  # Readout type for DPT decoder
+          FROM_PRETRAINED: "highlight_decoder.pt"  # Path to pretrained decoder weights (optional)  
+          USE_BN: False  # Use batch normalization in decoder
+          DROPOUT: 0.1  # Dropout rate in decoder layers
+          OUTPUT_IMAGE_SIZE: [896,896]  # Output image resolution [height, width]
+          OUTPUT_CHANNELS: 1  # Number of output channels (1 for highlight mask)
+          DECODER_LR: 0.0 #1.0e-5  # Custom learning rate for decoder (0.0 = frozen, 1.0 = same as base LR)
+          NUM_FUSION_BLOCKS_TRAINABLE: null  # Number of fusion blocks to train (0-4, null = train all if DECODER_LR != 0)
+      TOKEN_INPAINTER:
+        TOKEN_INPAINTER_CLASS: "TokenInpainter_Prior"  # Token inpainter class name
+        TOKEN_INPAINTER_MODULE: "token_inpainters"  # Module name to import token inpainter from
+        # FROM_PRETRAINED: "token_inpainter.pth"  # Path to pretrained token inpainter weights (optional)
+        TOKEN_INPAINTER_LR: 1.0e-4  # Learning rate for token inpainter (can differ from base LR)
+        DEPTH: 6  # Number of transformer blocks
+        HEADS: 16  # Number of attention heads
+        DROP: 0.05 # Dropout rate 
+        USE_POSITIONAL_ENCODING: True  # Enable 2D sinusoidal positional encodings
+        USE_FINAL_NORM: True  # Enable final LayerNorm before output projection
+        USE_LOCAL_PRIOR: True  # Blend local mean prior for masked seeds
+        LOCAL_PRIOR_WEIGHT: 0.2  # Weight for local prior blending (1.0 = only mask_token, 0.0 = only local mean)
+        LOCAL_PRIOR_KERNEL: 5  # Kernel size for local prior blending (> 1)
+        SEED_NOISE_STD: 0.02  # Standard deviation of noise added to masked seeds during training  
+
+  # FORWARD PASS PARAMETERS
+  INPAINT_MASK_THRESHOLD: 
+    value: 0.2  # Threshold for inpaint mask
+  INPAINT_MASK_DILATION: 
+    value: 40 # Dilation kernel size (pixels) for inpaint mask - Must be odd
+
+  USE_TORCH_COMPILE:  # Enable PyTorch 2.0 torch.compile for faster training (experimental)
+    value: False
+  DISTRIBUTE:
+    value: "ddp"
+
+  ### DATA                                                                      
+  DATASETS:
+    value:
+      # Reserved key: key-value pairs here override the same keys for every dataset (per-dataset entries still override this).
+      SCRREAM:
+        VAL_SCENES: ["scene10", "scene04"]  # List of validation scene names
+        RESIZE_MODE: "resize+crop"  # Image resizing mode: "resize", "crop", "resize+crop", or "pad"
+        SAMPLE_EVERY_N: 6  # Load every Nth frame from each scene (1 = all frames, 4 = every 4th frame)
+        
+      HOUSECAT6D:
+        VAL_SCENES: ["val_scene1","val_scene2"]  # Validation scene names
+        RESIZE_MODE: "resize+crop"  # Image resizing mode
+        SAMPLE_EVERY_N: 4  # Load every Nth frame
+
+      CROMO:
+        TRAIN_SCENES: ["kitchen"]  # Training scene names (list or string)
+        # VAL_SCENES: "station"  # Validation scene names (optional)
+        RESIZE_MODE: "resize"  # Image resizing mode
+        SAMPLE_EVERY_N: 2  # Load every Nth frame
+    
+      PSD:
+        TRAIN_SCENES: "PSD_Train"  # Training scene name (string or list)
+        VAL_SCENES: "PSD_Val"  # Validation scene name (string or list)
+        RESIZE_MODE: "resize+crop"  # Image resizing mode
+        SAMPLE_EVERY_N: 1  # Load every Nth frame (1 = all frames)
+
+      SCARED:
+        VAL_SCENES: ["v22","v23","v24","v25","v26","v27","v28","v29","v30","v31","v32","v33","v34"]  # Validation scene names
+        RESIZE_MODE: "resize+crop"  # Image resizing mode
+        SAMPLE_EVERY_N: 8  # Load every Nth frame
+              
+      STEREOMIS_TRACKING:
+        VAL_SCENES: ["P2_2"]  # Validation scene names
+        RESIZE_MODE: "resize+crop"  # Image resizing mode
+        SAMPLE_EVERY_N: 2  # Load every Nth frame
+
+      CHOLEC80:
+        TRAIN_SCENES: ["train"]  # Validation scene names
+        VAL_SCENES: ["test"]  # Validation scene names
+        RESIZE_MODE: "resize+crop"  # Image resizing mode
+        SAMPLE_EVERY_N: 40  # Load every Nth frame
+
+      SUNRGBD:
+        VAL_SCENES: ["realsense"]  # Validation scene names
+        RESIZE_MODE: "resize+crop"  # Image resizing mode
+        SAMPLE_EVERY_N: 4  # Load every Nth frame
+
+      SCANNET:
+        TRAIN_SCENES: ["train"]
+        VAL_SCENES: ["val"]
+        RESIZE_MODE: "resize+crop"
+        SAMPLE_EVERY_N: 5
+
+      OPENIMAGESV7:
+        TRAIN_SCENES: ["thescene"]
+        # VAL_SCENES: [""]
+        RESIZE_MODE: "resize+crop"
+        SAMPLE_EVERY_N: 5
+      
+      ENDOSYNTH:
+        TRAIN_SCENES: ["scene"]
+        # VAL_SCENES: ["val"]
+        RESIZE_MODE: "resize+crop"
+        SAMPLE_EVERY_N: 1
+
+      ALL_DATASETS:
+        FEW_IMAGES: False   # Override FEW_IMAGES for all datasets (for quick debugging set True)
+        TARGET_SIZE: [896,896]  # Override target image size [height, width] for all datasets
+        LOAD_RGB_ONLY: True        
+
+  BATCH_SIZE: # Max batch size with img size 896 is 32
+    value: 4    # Number of samples per batch (adjust based on GPU memory)
+  NUM_WORKERS:
+    value: 12  # Number of data loading worker processes (0 = main process only, "auto" = 90% of CPU affinity)
+  SHUFFLE:
+    value: True  # Shuffle training data each epoch (False for validation/test)
+  PIN_MEMORY:
+    value: True  # Pin memory in DataLoader for faster GPU transfer (recommended: True)
+  PREFETCH_FACTOR:
+    value: 2  # Number of batches to prefetch per worker (higher = more memory usage)
+  
+  ### HIGHLIGHTS                                                        
+  MOGE_MODEL:
+    value: "Ruicheng/moge-2-vits-normal"  # MoGe model name for normal estimation (HuggingFace format)
+  SURFACE_ROUGHNESS:
+    value: [100.0, 2000.0]  # Blinn-Phong surface roughness exponent (higher = sharper highlights)
+  INTENSITY:
+    value: [0.5, 1.0]  # Specular highlight intensity multiplier
+  LIGHT_DISTANCE_RANGE:
+    value: [0.0, 1]  # Range for light source distance sampling [min, max] (normalized)
+  LIGHT_LEFT_RIGHT_ANGLE:
+    value: [0, 360]  # Range for light source horizontal angle [min, max] in degrees
+  LIGHT_ABOVE_BELOW_ANGLE:
+    value: [0, 360]  # Range for light source vertical angle [min, max] in degrees
+  DATASET_HIGHLIGHT_DILATION:
+    value: 25  # Dilation kernel size (pixels) for dataset highlight masks
+  DATASET_HIGHLIGHT_THRESHOLD:
+    value: 0.85  # Brightness/luminance threshold (0-1) for detecting highlights in dataset images
+  DATASET_HIGHLIGHT_USE_LUMINANCE:
+    value: True  # If True, use perceptually-weighted luminance (0.299*R + 0.587*G + 0.114*B) for dataset highlights; if False, use simple mean brightness
+  HIGHLIGHT_COLOR:
+    value: [1.0, 1.0, 1.0]  # RGB color for synthetic highlights (normalized 0-1)
+  CLAMP_RECONSTRUCTION:
+    value: True  # Clamp reconstructed images to [0, 1] range if True
+
+  ### OPTIMIZATION                                                                      
+  LEARNING_RATE:
+    value: 1.0e-3  # Base learning rate for optimizer   # DEPRECATED. SETTING LR FOR EACH MODULE ABOVE
+  WEIGHT_DECAY:   
+    value: 0.0 # L2 regularization weight (0.0 = no weight decay)
+  EPOCHS:
+    value: 10  # Maximum number of training epochs
+  GRADIENT_ACCUMULATION_STEPS:
+    value: 1  # Number of steps to accumulate gradients before optimizer step (1 = no accumulation)
+  WARMUP:
+    value: 100  # Number of warmup steps for learning rate schedule (linear warmup from 0 to LR)
+  GRADIENT_CLIPPING_MAX_NORM:
+    value: 8  # Maximum gradient norm for clipping (set to -1 to disable clipping)
+  LR_SCHEDULER: 
+    value: 
+      ONPLATEAU:  # ReduceLROnPlateau scheduler (reduces LR when validation metric plateaus)
+        PATIENCE: 5  # Number of epochs to wait before reducing LR
+        FACTOR: 0.1  # Factor by which LR is reduced (new_lr = old_lr * factor)
+      COSINE:  # CosineAnnealingLR scheduler (cosine annealing schedule)
+        N_PERIODS: 0.5  # Number of cosine periods over training
+      # STEPWISE:  # StepLR scheduler (reduces LR at fixed step intervals)
+      #   N_STEPS: 4  # Number of times to reduce LR during training
+      #   GAMMA: 0.5  # Factor by which LR is reduced at each step (new_lr = old_lr * gamma)
+      # EXPONENTIAL:  # ExponentialLR scheduler (exponential decay)
+      #   GAMMA: 0.5  # Multiplicative factor for exponential decay
+
+  OPTIMIZER_BOOTSTRAP_NAME:
+    value: "AdamW"  # Optimizer name for initial training phase ("Adam", "SGD", etc.)
+  EARLY_STOPPING_PATIENCE:
+    value: 10  # Number of epochs without improvement before stopping training
+  SAVE_INTERVAL:
+    value: 1000  # Number of training steps between model checkpoints
+  
+  DATASET_HIGHLIGHT_SUPERVISION_THRESHOLD:
+    value: 0.1  # Pixel highlights above this threshold (should be low) are excluded from supervision
+
+  ### LOSS WEIGHTS (relative to the total loss, NOT NORMALIZED LATER)                                                                              
+  SPECULAR_LOSS_WEIGHT:
+    value: 0.0  # Weight for specular component reconstruction loss
+  DIFFUSE_LOSS_WEIGHT:
+    value: 1.0  # Weight for diffuse component reconstruction loss
+  HIGHLIGHT_LOSS_WEIGHT:
+    value: 1.0  # Weight for highlight mask regression loss
+  TOKEN_INPAINT_LOSS_WEIGHT:
+    value: 1.0  # Weight for token-space inpainting loss (L1 + cosine similarity in feature space)
+  IMAGE_RECONSTRUCTION_LOSS_WEIGHT:
+    value: 0.0  # Weight for full image reconstruction loss
+
+  HLREG_W_L1:
+    value: 1.0  # Weight for L1 loss in highlight regression
+  HLREG_USE_CHARB:
+    value: True  # Use Charbonnier loss (smooth L1) instead of standard L1 if True
+  HLREG_W_DICE:
+    value: 0.2  # Weight for Dice loss in highlight regression (for mask overlap)
+  HLREG_W_SSIM:
+    value: 0.0  # Weight for SSIM loss in highlight regression
+  HLREG_W_GRAD:
+    value: 0.0  # Weight for gradient loss in highlight regression
+  HLREG_W_TV:
+    value: 0.0  # Weight for total variation loss in highlight regression
+  HLREG_BALANCE_MODE:
+    value: "auto"   # Class balancing mode for highlight regression: 'none' | 'auto' | 'pos_weight'
+  HLREG_POS_WEIGHT:
+    value: 1.0      # Positive class weight (used only if BALANCE_MODE == 'pos_weight')
+  HLREG_FOCAL_GAMMA:
+    value: 2.0      # Focal loss gamma parameter (0.0 = standard BCE, 1.0-2.0 helps with gradient vanishing)
+  
+  WEIGHT_SEAM:
+    value: 0.5  # Weight for gradient matching loss on saturation ring
+  RING_DILATE_KERNEL:
+    value: 17  # Dilation kernel size (odd number) for creating ring mask around highlights
+  SEAM_USE_CHARB:
+    value: True  # Use Charbonnier loss instead of L1 in seam loss (smooth L1 for boundary consistency)
+  SEAM_WEIGHT_GRAD:
+    value: 0.0  # Weight for gradient matching term inside seam loss (0.0 = disable gradient term)
+  TOKEN_FEAT_ALPHA:
+    value: 0.5  # Mixing factor for token feature loss: alpha * L1 + (1-alpha) * (1-cosine_sim)
+
+  ### DIFFUSE HIGHLIGHT PENALTY
+  WEIGHT_DIFFUSE_HIGHLIGHT_PENALTY:
+    value: 0.75 # Weight for penalty loss on highlights in diffuse decoder output (0.0 = disabled)
+  DIFFUSE_HL_THRESHOLD:
+    value: 0.8  # Brightness/luminance threshold for detecting highlights in diffuse (0.0-1.0)
+  DIFFUSE_HL_USE_CHARB:
+    value: True  # Use Charbonnier loss instead of L1 for diffuse highlight penalty
+  DIFFUSE_HL_PENALTY_MODE:
+    value: "brightness"  # Penalty mode: "brightness" (penalize brightness/luminance above threshold) or "pixel" (penalize RGB values directly)
+  DIFFUSE_HL_TARGET_BRIGHTNESS:
+    value: null  # Target brightness/luminance for penalized pixels (null = use threshold value)
+  DIFFUSE_HL_USE_LUMINANCE:
+    value: False  # If True, use perceptually-weighted luminance (0.299*R + 0.587*G + 0.114*B); if False, use simple mean brightness
+
+  ### LOGGING, RESULTS AND WANDB                                                                 
+  LOG_INTERVAL:
+    value: 1  # Number of training steps between console log outputs
+  WANDB_LOG_INTERVAL:
+    value: 1  # Number of training steps between WandB metric logs
+  IMAGE_LOG_INTERVAL:
+    value: 5  # Number of training steps between image logging to WandB
+  NO_WANDB:
+    value: False  # Disable WandB logging if True (useful for local debugging)
+  MODEL_WATCHER_FREQ_WANDB:
+    value: 50  # Frequency (in steps) for logging model parameter histograms to WandB
+  WANDB_ENTITY:
+    value: "unreflect-anything"  # WandB organization/entity name
+  WANDB_PROJECT:
+    value: "UnReflectAnything"  # WandB project name
+  NOTES:
+    value: "LOCAL_PRIOR_WEIGHT 0.2"  # Notes/description for this training run

configs/{end2end.yaml → end2end08.yaml}

@@ -36,7 +36,7 @@ parameters:
           DROPOUT: 0.1  # Dropout rate in decoder layers
           OUTPUT_IMAGE_SIZE: [896,896]  # Output image resolution [height, width]
           OUTPUT_CHANNELS: 1  # Number of output channels (1 for highlight mask)
-          DECODER_LR: 1.0e-5  # Custom learning rate for decoder (0.0 = frozen, 1.0 = same as base LR)
+          DECODER_LR: 0.0 #1.0e-5  # Custom learning rate for decoder (0.0 = frozen, 1.0 = same as base LR)
           NUM_FUSION_BLOCKS_TRAINABLE: null  # Number of fusion blocks to train (0-4, null = train all if DECODER_LR != 0)
       TOKEN_INPAINTER:
         TOKEN_INPAINTER_CLASS: "TokenInpainter_Prior"  # Token inpainter class name
@@ -135,7 +135,7 @@ parameters:
         LOAD_RGB_ONLY: True        
 
   BATCH_SIZE: # Max batch size with img size 896 is 32
-    value: 4 # Number of samples per batch (adjust based on GPU memory)
+    value: 4    # Number of samples per batch (adjust based on GPU memory)
   NUM_WORKERS:
     value: 12  # Number of data loading worker processes (0 = main process only, "auto" = 90% of CPU affinity)
   SHUFFLE:
@@ -149,9 +149,9 @@ parameters:
   MOGE_MODEL:
     value: "Ruicheng/moge-2-vits-normal"  # MoGe model name for normal estimation (HuggingFace format)
   SURFACE_ROUGHNESS:
-    value: 100.0  # Blinn-Phong surface roughness exponent (higher = sharper highlights)
+    value: [100.0, 2000.0]  # Blinn-Phong surface roughness exponent (higher = sharper highlights)
   INTENSITY:
-    value: 0.8  # Specular highlight intensity multiplier
+    value: [0.5, 1.0]  # Specular highlight intensity multiplier
   LIGHT_DISTANCE_RANGE:
     value: [0.0, 1]  # Range for light source distance sampling [min, max] (normalized)
   LIGHT_LEFT_RIGHT_ANGLE:
@@ -277,4 +277,4 @@ parameters:
   WANDB_PROJECT:
     value: "UnReflectAnything"  # WandB project name
   NOTES:
-    value: "LOCAL_PRIOR_WEIGHT 0.9"  # Notes/description for this training run
+    value: "LOCAL_PRIOR_WEIGHT 0.8"  # Notes/description for this training run

configs/highlight_decoder_pretrain.yaml

@@ -41,7 +41,7 @@ parameters:
         LOCAL_PRIOR_KERNEL: 5  # Kernel size for local prior blending (> 1)
         SEED_NOISE_STD: 0.02  # Standard deviation of noise added to masked seeds during training  
   INPAINT_MASK_DILATION:
-    value: None  # Dilation kernel size (pixels) for inpaint mask (None = compute based on image size)
+    value: 40  # Dilation kernel size (pixels) for inpaint mask (None = compute based on image size)
   INPAINT_MASK_THRESHOLD:
     value: 0.2  # Inpaint selection threshold
   USE_TORCH_COMPILE:  # Enable PyTorch 2.0 torch.compile for faster training (experimental)

configs/pretrained_config.yaml

@@ -0,0 +1,54 @@
+### BASELINE: CONVERGES AFTER LONG
+
+parameters:
+
+  ### MODEL ARCHITECTURE                                                       
+  MODEL:
+    value:
+      MODEL_CLASS: "UnReflect_Model_TokenInpainter"  # Main model class name (must match class in models.py)
+      MODEL_MODULE: "models"  # Module name to import model classes from (default: "models")
+      RGB_ENCODER:
+        ENCODER: "facebook/dinov3-vitl16-pretrain-lvd1689m"  # DINOv3 encoder model name (HuggingFace format)
+        IMAGE_SIZE: 896  # Input image size (height and width in pixels)
+        RETURN_SELECTED_LAYERS: [3, 6, 9, 12]  # Transformer layer indices to extract features from (0-indexed)
+        RGB_ENCODER_LR: 0.0  # Learning rate for RGB encoder (0.0 = frozen, must be explicitly set)
+      DECODERS:
+        diffuse:
+          FEATURE_DIM: 1024  # Feature dimension for decoder (should match encoder output)
+          REASSEMBLE_OUT_CHANNELS: [768,1024,1536,2048]  # Output channels for each decoder stage (DPT-style reassembly)
+          REASSEMBLE_FACTORS: [4.0, 2.0, 1.0, 0.5]  # Spatial upsampling factors for each stage
+          READOUT_TYPE: "ignore"  # Readout type for DPT decoder ("ignore", "project", etc.)
+          # FROM_PRETRAINED: "diffuse_decoder.pt"  # Path to pretrained decoder weights (optional)
+          USE_BN: False  # Use batch normalization in decoder
+          DROPOUT: 0.1  # Dropout rate in decoder layers
+          OUTPUT_IMAGE_SIZE: [896,896]  # Output image resolution [height, width]
+          OUTPUT_CHANNELS: 3  # Number of output channels (3 for RGB diffuse image)
+          DECODER_LR: 0.0 # Custom learning rate for decoder (0.0 = frozen, 1.0 = same as base LR)
+          NUM_FUSION_BLOCKS_TRAINABLE: 1  # Number of fusion blocks to train (0-4, null = train all if DECODER_LR != 0)
+          TRAIN_RGB_HEAD: True  # Whether to train RGB head (true/false, null = train if DECODER_LR != 0)
+        highlight:
+          FEATURE_DIM: 1024  # Feature dimension for highlight decoder
+          REASSEMBLE_OUT_CHANNELS: [96,192,384,768]  # Output channels for each decoder stage
+          REASSEMBLE_FACTORS: [4.0, 2.0, 1.0, 0.5]  # Spatial upsampling factors for each stage
+          READOUT_TYPE: "ignore"  # Readout type for DPT decoder
+          # FROM_PRETRAINED: "highlight_decoder.pt"  # Path to pretrained decoder weights (optional)
+          USE_BN: False  # Use batch normalization in decoder
+          DROPOUT: 0.1  # Dropout rate in decoder layers
+          OUTPUT_IMAGE_SIZE: [896,896]  # Output image resolution [height, width]
+          OUTPUT_CHANNELS: 1  # Number of output channels (1 for highlight mask)
+          DECODER_LR: 5.0e-4  # Custom learning rate for decoder (0.0 = frozen, 1.0 = same as base LR)
+          NUM_FUSION_BLOCKS_TRAINABLE: null  # Number of fusion blocks to train (0-4, null = train all if DECODER_LR != 0)
+      TOKEN_INPAINTER:
+        TOKEN_INPAINTER_CLASS: "TokenInpainter_Prior"  # Token inpainter class name
+        TOKEN_INPAINTER_MODULE: "token_inpainters"  # Module name to import token inpainter from
+        # FROM_PRETRAINED: "token_inpainter.pth"  # Path to pretrained token inpainter weights (optional)
+        TOKEN_INPAINTER_LR: 1.0e-4  # Learning rate for token inpainter (can differ from base LR)
+        DEPTH: 6  # Number of transformer blocks
+        HEADS: 16  # Number of attention heads
+        DROP: 0.05 # Dropout rate 
+        USE_POSITIONAL_ENCODING: True  # Enable 2D sinusoidal positional encodings
+        USE_FINAL_NORM: True  # Enable final LayerNorm before output projection
+        USE_LOCAL_PRIOR: True  # Blend local mean prior for masked seeds
+        LOCAL_PRIOR_WEIGHT: 0.5  # Weight for local prior blending (1.0 = only mask_token, 0.0 = only local mean)
+        LOCAL_PRIOR_KERNEL: 5  # Kernel size for local prior blending (> 1)
+        SEED_NOISE_STD: 0.02  # Standard deviation of noise added to masked seeds during training