wavedl 1.5.7__py3-none-any.whl → 1.6.1__py3-none-any.whl

wavedl/models/regnet.py

@@ -1,406 +1,406 @@
-"""
-RegNet: Designing Network Design Spaces
-========================================
-
-RegNet provides a family of models with predictable scaling behavior,
-designed through systematic exploration of network design spaces.
-Models scale smoothly from mobile to server deployments.
-
-**Key Features**:
-    - Predictable scaling: accuracy increases linearly with compute
-    - Simple, uniform architecture (no complex compound scaling)
-    - Group convolutions for efficiency
-    - Optional Squeeze-and-Excitation (SE) attention
-
-**Variants** (RegNetY includes SE attention):
-    - regnet_y_400mf: Ultra-light (~4.0M params, 0.4 GFLOPs)
-    - regnet_y_800mf: Light (~5.8M params, 0.8 GFLOPs)
-    - regnet_y_1_6gf: Medium (~10.5M params, 1.6 GFLOPs) - Recommended
-    - regnet_y_3_2gf: Large (~18.3M params, 3.2 GFLOPs)
-    - regnet_y_8gf: Very large (~37.9M params, 8.0 GFLOPs)
-
-**When to Use RegNet**:
-    - When you need predictable performance at a given compute budget
-    - For systematic model selection experiments
-    - When interpretability of design choices matters
-    - As an efficient alternative to ResNet
-
-**Note**: RegNet is 2D-only. For 1D data, use TCN. For 3D data, use ResNet3D.
-
-References:
-    Radosavovic, I., et al. (2020). Designing Network Design Spaces.
-    CVPR 2020. https://arxiv.org/abs/2003.13678
-
-Author: Ductho Le (ductho.le@outlook.com)
-"""
-
-from typing import Any
-
-import torch
-import torch.nn as nn
-
-
-try:
-    from torchvision.models import (
-        RegNet_Y_1_6GF_Weights,
-        RegNet_Y_3_2GF_Weights,
-        RegNet_Y_8GF_Weights,
-        RegNet_Y_400MF_Weights,
-        RegNet_Y_800MF_Weights,
-        regnet_y_1_6gf,
-        regnet_y_3_2gf,
-        regnet_y_8gf,
-        regnet_y_400mf,
-        regnet_y_800mf,
-    )
-
-    REGNET_AVAILABLE = True
-except ImportError:
-    REGNET_AVAILABLE = False
-
-from wavedl.models.base import BaseModel
-from wavedl.models.registry import register_model
-
-
-class RegNetBase(BaseModel):
-    """
-    Base RegNet class for regression tasks.
-
-    Wraps torchvision RegNetY (with SE attention) with:
-    - Optional pretrained weights (ImageNet-1K)
-    - Automatic input channel adaptation (grayscale → 3ch)
-    - Custom regression head
-
-    RegNet advantages:
-    - Simple, uniform design (easy to understand and modify)
-    - Predictable accuracy/compute trade-off
-    - Efficient group convolutions
-    - SE attention for channel weighting (RegNetY variants)
-
-    Note: This is 2D-only. Input shape must be (H, W).
-    """
-
-    def __init__(
-        self,
-        in_shape: tuple[int, int],
-        out_size: int,
-        model_fn,
-        weights_class,
-        pretrained: bool = True,
-        dropout_rate: float = 0.2,
-        freeze_backbone: bool = False,
-        regression_hidden: int = 256,
-        **kwargs,
-    ):
-        """
-        Initialize RegNet for regression.
-
-        Args:
-            in_shape: (H, W) input image dimensions
-            out_size: Number of regression output targets
-            model_fn: torchvision model constructor
-            weights_class: Pretrained weights enum class
-            pretrained: Use ImageNet pretrained weights (default: True)
-            dropout_rate: Dropout rate in regression head (default: 0.2)
-            freeze_backbone: Freeze backbone for fine-tuning (default: False)
-            regression_hidden: Hidden units in regression head (default: 256)
-        """
-        super().__init__(in_shape, out_size)
-
-        if not REGNET_AVAILABLE:
-            raise ImportError(
-                "torchvision is required for RegNet. "
-                "Install with: pip install torchvision"
-            )
-
-        if len(in_shape) != 2:
-            raise ValueError(
-                f"RegNet requires 2D input (H, W), got {len(in_shape)}D. "
-                "For 1D data, use TCN. For 3D data, use ResNet3D."
-            )
-
-        self.pretrained = pretrained
-        self.dropout_rate = dropout_rate
-        self.freeze_backbone = freeze_backbone
-        self.regression_hidden = regression_hidden
-
-        # Load pretrained backbone
-        weights = weights_class.IMAGENET1K_V1 if pretrained else None
-        self.backbone = model_fn(weights=weights)
-
-        # RegNet uses .fc as the classification head
-        in_features = self.backbone.fc.in_features
-
-        # Replace fc with regression head
-        self.backbone.fc = nn.Sequential(
-            nn.Dropout(dropout_rate),
-            nn.Linear(in_features, regression_hidden),
-            nn.ReLU(inplace=True),
-            nn.Dropout(dropout_rate * 0.5),
-            nn.Linear(regression_hidden, out_size),
-        )
-
-        # Adapt first conv for single-channel input (3× memory savings vs expand)
-        self._adapt_input_channels()
-
-        # Optionally freeze backbone for fine-tuning (after adaptation so new conv is frozen too)
-        if freeze_backbone:
-            self._freeze_backbone()
-
-    def _adapt_input_channels(self):
-        """Modify first conv to accept single-channel input.
-
-        Instead of expanding 1→3 channels in forward (which triples memory),
-        we replace the first conv layer with a 1-channel version and initialize
-        weights as the mean of the pretrained RGB filters.
-        """
-        old_conv = self.backbone.stem[0]
-        new_conv = nn.Conv2d(
-            1,  # Single channel input
-            old_conv.out_channels,
-            kernel_size=old_conv.kernel_size,
-            stride=old_conv.stride,
-            padding=old_conv.padding,
-            dilation=old_conv.dilation,
-            groups=old_conv.groups,
-            padding_mode=old_conv.padding_mode,
-            bias=old_conv.bias is not None,
-        )
-        if self.pretrained:
-            with torch.no_grad():
-                new_conv.weight.copy_(old_conv.weight.mean(dim=1, keepdim=True))
-        self.backbone.stem[0] = new_conv
-
-    def _freeze_backbone(self):
-        """Freeze all backbone parameters except the fc layer."""
-        for name, param in self.backbone.named_parameters():
-            if "fc" not in name:
-                param.requires_grad = False
-
-    def forward(self, x: torch.Tensor) -> torch.Tensor:
-        """
-        Forward pass.
-
-        Args:
-            x: Input tensor of shape (B, 1, H, W)
-
-        Returns:
-            Output tensor of shape (B, out_size)
-        """
-        return self.backbone(x)
-
-    @classmethod
-    def get_default_config(cls) -> dict[str, Any]:
-        """Return default configuration for RegNet."""
-        return {
-            "pretrained": True,
-            "dropout_rate": 0.2,
-            "freeze_backbone": False,
-            "regression_hidden": 256,
-        }
-
-
-# =============================================================================
-# REGISTERED MODEL VARIANTS
-# =============================================================================
-
-
-@register_model("regnet_y_400mf")
-class RegNetY400MF(RegNetBase):
-    """
-    RegNetY-400MF: Ultra-lightweight for constrained environments.
-
-    ~4.0M parameters, 0.4 GFLOPs. Smallest RegNet variant with SE attention.
-
-    Recommended for:
-        - Edge deployment with moderate accuracy needs
-        - Quick training experiments
-        - Baseline comparisons
-
-    Args:
-        in_shape: (H, W) image dimensions
-        out_size: Number of regression targets
-        pretrained: Use ImageNet pretrained weights (default: True)
-        dropout_rate: Dropout rate in head (default: 0.2)
-        freeze_backbone: Freeze backbone for fine-tuning (default: False)
-        regression_hidden: Hidden units in regression head (default: 256)
-
-    Example:
-        >>> model = RegNetY400MF(in_shape=(224, 224), out_size=3)
-        >>> x = torch.randn(4, 1, 224, 224)
-        >>> out = model(x)  # (4, 3)
-    """
-
-    def __init__(self, in_shape: tuple[int, int], out_size: int, **kwargs):
-        super().__init__(
-            in_shape=in_shape,
-            out_size=out_size,
-            model_fn=regnet_y_400mf,
-            weights_class=RegNet_Y_400MF_Weights,
-            **kwargs,
-        )
-
-    def __repr__(self) -> str:
-        pt = "pretrained" if self.pretrained else "scratch"
-        return f"RegNetY_400MF({pt}, in={self.in_shape}, out={self.out_size})"
-
-
-@register_model("regnet_y_800mf")
-class RegNetY800MF(RegNetBase):
-    """
-    RegNetY-800MF: Light variant with good accuracy.
-
-    ~6.4M parameters, 0.8 GFLOPs. Good balance for mobile deployment.
-
-    Recommended for:
-        - Mobile/portable devices
-        - When MobileNet isn't accurate enough
-        - Moderate compute budgets
-
-    Args:
-        in_shape: (H, W) image dimensions
-        out_size: Number of regression targets
-        pretrained: Use ImageNet pretrained weights (default: True)
-        dropout_rate: Dropout rate in head (default: 0.2)
-        freeze_backbone: Freeze backbone for fine-tuning (default: False)
-        regression_hidden: Hidden units in regression head (default: 256)
-
-    Example:
-        >>> model = RegNetY800MF(in_shape=(224, 224), out_size=3)
-        >>> x = torch.randn(4, 1, 224, 224)
-        >>> out = model(x)  # (4, 3)
-    """
-
-    def __init__(self, in_shape: tuple[int, int], out_size: int, **kwargs):
-        super().__init__(
-            in_shape=in_shape,
-            out_size=out_size,
-            model_fn=regnet_y_800mf,
-            weights_class=RegNet_Y_800MF_Weights,
-            **kwargs,
-        )
-
-    def __repr__(self) -> str:
-        pt = "pretrained" if self.pretrained else "scratch"
-        return f"RegNetY_800MF({pt}, in={self.in_shape}, out={self.out_size})"
-
-
-@register_model("regnet_y_1_6gf")
-class RegNetY1_6GF(RegNetBase):
-    """
-    RegNetY-1.6GF: Recommended default for balanced performance.
-
-    ~11.2M parameters, 1.6 GFLOPs. Best trade-off of accuracy and efficiency.
-    Comparable to ResNet50 but more efficient.
-
-    Recommended for:
-        - Default choice for general wave-based tasks
-        - When you want predictable scaling
-        - Server deployment with efficiency needs
-
-    Args:
-        in_shape: (H, W) image dimensions
-        out_size: Number of regression targets
-        pretrained: Use ImageNet pretrained weights (default: True)
-        dropout_rate: Dropout rate in head (default: 0.2)
-        freeze_backbone: Freeze backbone for fine-tuning (default: False)
-        regression_hidden: Hidden units in regression head (default: 256)
-
-    Example:
-        >>> model = RegNetY1_6GF(in_shape=(224, 224), out_size=3)
-        >>> x = torch.randn(4, 1, 224, 224)
-        >>> out = model(x)  # (4, 3)
-    """
-
-    def __init__(self, in_shape: tuple[int, int], out_size: int, **kwargs):
-        super().__init__(
-            in_shape=in_shape,
-            out_size=out_size,
-            model_fn=regnet_y_1_6gf,
-            weights_class=RegNet_Y_1_6GF_Weights,
-            **kwargs,
-        )
-
-    def __repr__(self) -> str:
-        pt = "pretrained" if self.pretrained else "scratch"
-        return f"RegNetY_1.6GF({pt}, in={self.in_shape}, out={self.out_size})"
-
-
-@register_model("regnet_y_3_2gf")
-class RegNetY3_2GF(RegNetBase):
-    """
-    RegNetY-3.2GF: Higher accuracy for demanding tasks.
-
-    ~19.4M parameters, 3.2 GFLOPs. Use when 1.6GF isn't sufficient.
-
-    Recommended for:
-        - Larger datasets requiring more capacity
-        - When accuracy is more important than efficiency
-        - Research experiments with multiple model sizes
-
-    Args:
-        in_shape: (H, W) image dimensions
-        out_size: Number of regression targets
-        pretrained: Use ImageNet pretrained weights (default: True)
-        dropout_rate: Dropout rate in head (default: 0.2)
-        freeze_backbone: Freeze backbone for fine-tuning (default: False)
-        regression_hidden: Hidden units in regression head (default: 256)
-
-    Example:
-        >>> model = RegNetY3_2GF(in_shape=(224, 224), out_size=3)
-        >>> x = torch.randn(4, 1, 224, 224)
-        >>> out = model(x)  # (4, 3)
-    """
-
-    def __init__(self, in_shape: tuple[int, int], out_size: int, **kwargs):
-        super().__init__(
-            in_shape=in_shape,
-            out_size=out_size,
-            model_fn=regnet_y_3_2gf,
-            weights_class=RegNet_Y_3_2GF_Weights,
-            **kwargs,
-        )
-
-    def __repr__(self) -> str:
-        pt = "pretrained" if self.pretrained else "scratch"
-        return f"RegNetY_3.2GF({pt}, in={self.in_shape}, out={self.out_size})"
-
-
-@register_model("regnet_y_8gf")
-class RegNetY8GF(RegNetBase):
-    """
-    RegNetY-8GF: High capacity for large-scale tasks.
-
-    ~39.2M parameters, 8.0 GFLOPs. Use for maximum accuracy needs.
-
-    Recommended for:
-        - Very large datasets (>50k samples)
-        - Complex wave patterns
-        - HPC environments with ample GPU memory
-
-    Args:
-        in_shape: (H, W) image dimensions
-        out_size: Number of regression targets
-        pretrained: Use ImageNet pretrained weights (default: True)
-        dropout_rate: Dropout rate in head (default: 0.2)
-        freeze_backbone: Freeze backbone for fine-tuning (default: False)
-        regression_hidden: Hidden units in regression head (default: 256)
-
-    Example:
-        >>> model = RegNetY8GF(in_shape=(224, 224), out_size=3)
-        >>> x = torch.randn(4, 1, 224, 224)
-        >>> out = model(x)  # (4, 3)
-    """
-
-    def __init__(self, in_shape: tuple[int, int], out_size: int, **kwargs):
-        super().__init__(
-            in_shape=in_shape,
-            out_size=out_size,
-            model_fn=regnet_y_8gf,
-            weights_class=RegNet_Y_8GF_Weights,
-            **kwargs,
-        )
-
-    def __repr__(self) -> str:
-        pt = "pretrained" if self.pretrained else "scratch"
-        return f"RegNetY_8GF({pt}, in={self.in_shape}, out={self.out_size})"
+"""
+RegNet: Designing Network Design Spaces
+========================================
+
+RegNet provides a family of models with predictable scaling behavior,
+designed through systematic exploration of network design spaces.
+Models scale smoothly from mobile to server deployments.
+
+**Key Features**:
+    - Predictable scaling: accuracy increases linearly with compute
+    - Simple, uniform architecture (no complex compound scaling)
+    - Group convolutions for efficiency
+    - Optional Squeeze-and-Excitation (SE) attention
+
+**Variants** (RegNetY includes SE attention):
+    - regnet_y_400mf: Ultra-light (~3.9M backbone params, 0.4 GFLOPs)
+    - regnet_y_800mf: Light (~5.7M backbone params, 0.8 GFLOPs)
+    - regnet_y_1_6gf: Medium (~10.3M backbone params, 1.6 GFLOPs) - Recommended
+    - regnet_y_3_2gf: Large (~17.9M backbone params, 3.2 GFLOPs)
+    - regnet_y_8gf: Very large (~37.4M backbone params, 8.0 GFLOPs)
+
+**When to Use RegNet**:
+    - When you need predictable performance at a given compute budget
+    - For systematic model selection experiments
+    - When interpretability of design choices matters
+    - As an efficient alternative to ResNet
+
+**Note**: RegNet is 2D-only. For 1D data, use TCN. For 3D data, use ResNet3D.
+
+References:
+    Radosavovic, I., et al. (2020). Designing Network Design Spaces.
+    CVPR 2020. https://arxiv.org/abs/2003.13678
+
+Author: Ductho Le (ductho.le@outlook.com)
+"""
+
+from typing import Any
+
+import torch
+import torch.nn as nn
+
+
+try:
+    from torchvision.models import (
+        RegNet_Y_1_6GF_Weights,
+        RegNet_Y_3_2GF_Weights,
+        RegNet_Y_8GF_Weights,
+        RegNet_Y_400MF_Weights,
+        RegNet_Y_800MF_Weights,
+        regnet_y_1_6gf,
+        regnet_y_3_2gf,
+        regnet_y_8gf,
+        regnet_y_400mf,
+        regnet_y_800mf,
+    )
+
+    REGNET_AVAILABLE = True
+except ImportError:
+    REGNET_AVAILABLE = False
+
+from wavedl.models.base import BaseModel
+from wavedl.models.registry import register_model
+
+
+class RegNetBase(BaseModel):
+    """
+    Base RegNet class for regression tasks.
+
+    Wraps torchvision RegNetY (with SE attention) with:
+    - Optional pretrained weights (ImageNet-1K)
+    - Automatic input channel adaptation (grayscale → 3ch)
+    - Custom regression head
+
+    RegNet advantages:
+    - Simple, uniform design (easy to understand and modify)
+    - Predictable accuracy/compute trade-off
+    - Efficient group convolutions
+    - SE attention for channel weighting (RegNetY variants)
+
+    Note: This is 2D-only. Input shape must be (H, W).
+    """
+
+    def __init__(
+        self,
+        in_shape: tuple[int, int],
+        out_size: int,
+        model_fn,
+        weights_class,
+        pretrained: bool = True,
+        dropout_rate: float = 0.2,
+        freeze_backbone: bool = False,
+        regression_hidden: int = 256,
+        **kwargs,
+    ):
+        """
+        Initialize RegNet for regression.
+
+        Args:
+            in_shape: (H, W) input image dimensions
+            out_size: Number of regression output targets
+            model_fn: torchvision model constructor
+            weights_class: Pretrained weights enum class
+            pretrained: Use ImageNet pretrained weights (default: True)
+            dropout_rate: Dropout rate in regression head (default: 0.2)
+            freeze_backbone: Freeze backbone for fine-tuning (default: False)
+            regression_hidden: Hidden units in regression head (default: 256)
+        """
+        super().__init__(in_shape, out_size)
+
+        if not REGNET_AVAILABLE:
+            raise ImportError(
+                "torchvision is required for RegNet. "
+                "Install with: pip install torchvision"
+            )
+
+        if len(in_shape) != 2:
+            raise ValueError(
+                f"RegNet requires 2D input (H, W), got {len(in_shape)}D. "
+                "For 1D data, use TCN. For 3D data, use ResNet3D."
+            )
+
+        self.pretrained = pretrained
+        self.dropout_rate = dropout_rate
+        self.freeze_backbone = freeze_backbone
+        self.regression_hidden = regression_hidden
+
+        # Load pretrained backbone
+        weights = weights_class.IMAGENET1K_V1 if pretrained else None
+        self.backbone = model_fn(weights=weights)
+
+        # RegNet uses .fc as the classification head
+        in_features = self.backbone.fc.in_features
+
+        # Replace fc with regression head
+        self.backbone.fc = nn.Sequential(
+            nn.Dropout(dropout_rate),
+            nn.Linear(in_features, regression_hidden),
+            nn.ReLU(inplace=True),
+            nn.Dropout(dropout_rate * 0.5),
+            nn.Linear(regression_hidden, out_size),
+        )
+
+        # Adapt first conv for single-channel input (3× memory savings vs expand)
+        self._adapt_input_channels()
+
+        # Optionally freeze backbone for fine-tuning (after adaptation so new conv is frozen too)
+        if freeze_backbone:
+            self._freeze_backbone()
+
+    def _adapt_input_channels(self):
+        """Modify first conv to accept single-channel input.
+
+        Instead of expanding 1→3 channels in forward (which triples memory),
+        we replace the first conv layer with a 1-channel version and initialize
+        weights as the mean of the pretrained RGB filters.
+        """
+        old_conv = self.backbone.stem[0]
+        new_conv = nn.Conv2d(
+            1,  # Single channel input
+            old_conv.out_channels,
+            kernel_size=old_conv.kernel_size,
+            stride=old_conv.stride,
+            padding=old_conv.padding,
+            dilation=old_conv.dilation,
+            groups=old_conv.groups,
+            padding_mode=old_conv.padding_mode,
+            bias=old_conv.bias is not None,
+        )
+        if self.pretrained:
+            with torch.no_grad():
+                new_conv.weight.copy_(old_conv.weight.mean(dim=1, keepdim=True))
+        self.backbone.stem[0] = new_conv
+
+    def _freeze_backbone(self):
+        """Freeze all backbone parameters except the fc layer."""
+        for name, param in self.backbone.named_parameters():
+            if "fc" not in name:
+                param.requires_grad = False
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        """
+        Forward pass.
+
+        Args:
+            x: Input tensor of shape (B, 1, H, W)
+
+        Returns:
+            Output tensor of shape (B, out_size)
+        """
+        return self.backbone(x)
+
+    @classmethod
+    def get_default_config(cls) -> dict[str, Any]:
+        """Return default configuration for RegNet."""
+        return {
+            "pretrained": True,
+            "dropout_rate": 0.2,
+            "freeze_backbone": False,
+            "regression_hidden": 256,
+        }
+
+
+# =============================================================================
+# REGISTERED MODEL VARIANTS
+# =============================================================================
+
+
+@register_model("regnet_y_400mf")
+class RegNetY400MF(RegNetBase):
+    """
+    RegNetY-400MF: Ultra-lightweight for constrained environments.
+
+    ~3.9M backbone parameters, 0.4 GFLOPs. Smallest RegNet variant with SE attention.
+
+    Recommended for:
+        - Edge deployment with moderate accuracy needs
+        - Quick training experiments
+        - Baseline comparisons
+
+    Args:
+        in_shape: (H, W) image dimensions
+        out_size: Number of regression targets
+        pretrained: Use ImageNet pretrained weights (default: True)
+        dropout_rate: Dropout rate in head (default: 0.2)
+        freeze_backbone: Freeze backbone for fine-tuning (default: False)
+        regression_hidden: Hidden units in regression head (default: 256)
+
+    Example:
+        >>> model = RegNetY400MF(in_shape=(224, 224), out_size=3)
+        >>> x = torch.randn(4, 1, 224, 224)
+        >>> out = model(x)  # (4, 3)
+    """
+
+    def __init__(self, in_shape: tuple[int, int], out_size: int, **kwargs):
+        super().__init__(
+            in_shape=in_shape,
+            out_size=out_size,
+            model_fn=regnet_y_400mf,
+            weights_class=RegNet_Y_400MF_Weights,
+            **kwargs,
+        )
+
+    def __repr__(self) -> str:
+        pt = "pretrained" if self.pretrained else "scratch"
+        return f"RegNetY_400MF({pt}, in={self.in_shape}, out={self.out_size})"
+
+
+@register_model("regnet_y_800mf")
+class RegNetY800MF(RegNetBase):
+    """
+    RegNetY-800MF: Light variant with good accuracy.
+
+    ~5.7M backbone parameters, 0.8 GFLOPs. Good balance for mobile deployment.
+
+    Recommended for:
+        - Mobile/portable devices
+        - When MobileNet isn't accurate enough
+        - Moderate compute budgets
+
+    Args:
+        in_shape: (H, W) image dimensions
+        out_size: Number of regression targets
+        pretrained: Use ImageNet pretrained weights (default: True)
+        dropout_rate: Dropout rate in head (default: 0.2)
+        freeze_backbone: Freeze backbone for fine-tuning (default: False)
+        regression_hidden: Hidden units in regression head (default: 256)
+
+    Example:
+        >>> model = RegNetY800MF(in_shape=(224, 224), out_size=3)
+        >>> x = torch.randn(4, 1, 224, 224)
+        >>> out = model(x)  # (4, 3)
+    """
+
+    def __init__(self, in_shape: tuple[int, int], out_size: int, **kwargs):
+        super().__init__(
+            in_shape=in_shape,
+            out_size=out_size,
+            model_fn=regnet_y_800mf,
+            weights_class=RegNet_Y_800MF_Weights,
+            **kwargs,
+        )
+
+    def __repr__(self) -> str:
+        pt = "pretrained" if self.pretrained else "scratch"
+        return f"RegNetY_800MF({pt}, in={self.in_shape}, out={self.out_size})"
+
+
+@register_model("regnet_y_1_6gf")
+class RegNetY1_6GF(RegNetBase):
+    """
+    RegNetY-1.6GF: Recommended default for balanced performance.
+
+    ~10.3M backbone parameters, 1.6 GFLOPs. Best trade-off of accuracy and efficiency.
+    Comparable to ResNet50 but more efficient.
+
+    Recommended for:
+        - Default choice for general wave-based tasks
+        - When you want predictable scaling
+        - Server deployment with efficiency needs
+
+    Args:
+        in_shape: (H, W) image dimensions
+        out_size: Number of regression targets
+        pretrained: Use ImageNet pretrained weights (default: True)
+        dropout_rate: Dropout rate in head (default: 0.2)
+        freeze_backbone: Freeze backbone for fine-tuning (default: False)
+        regression_hidden: Hidden units in regression head (default: 256)
+
+    Example:
+        >>> model = RegNetY1_6GF(in_shape=(224, 224), out_size=3)
+        >>> x = torch.randn(4, 1, 224, 224)
+        >>> out = model(x)  # (4, 3)
+    """
+
+    def __init__(self, in_shape: tuple[int, int], out_size: int, **kwargs):
+        super().__init__(
+            in_shape=in_shape,
+            out_size=out_size,
+            model_fn=regnet_y_1_6gf,
+            weights_class=RegNet_Y_1_6GF_Weights,
+            **kwargs,
+        )
+
+    def __repr__(self) -> str:
+        pt = "pretrained" if self.pretrained else "scratch"
+        return f"RegNetY_1.6GF({pt}, in={self.in_shape}, out={self.out_size})"
+
+
+@register_model("regnet_y_3_2gf")
+class RegNetY3_2GF(RegNetBase):
+    """
+    RegNetY-3.2GF: Higher accuracy for demanding tasks.
+
+    ~17.9M backbone parameters, 3.2 GFLOPs. Use when 1.6GF isn't sufficient.
+
+    Recommended for:
+        - Larger datasets requiring more capacity
+        - When accuracy is more important than efficiency
+        - Research experiments with multiple model sizes
+
+    Args:
+        in_shape: (H, W) image dimensions
+        out_size: Number of regression targets
+        pretrained: Use ImageNet pretrained weights (default: True)
+        dropout_rate: Dropout rate in head (default: 0.2)
+        freeze_backbone: Freeze backbone for fine-tuning (default: False)
+        regression_hidden: Hidden units in regression head (default: 256)
+
+    Example:
+        >>> model = RegNetY3_2GF(in_shape=(224, 224), out_size=3)
+        >>> x = torch.randn(4, 1, 224, 224)
+        >>> out = model(x)  # (4, 3)
+    """
+
+    def __init__(self, in_shape: tuple[int, int], out_size: int, **kwargs):
+        super().__init__(
+            in_shape=in_shape,
+            out_size=out_size,
+            model_fn=regnet_y_3_2gf,
+            weights_class=RegNet_Y_3_2GF_Weights,
+            **kwargs,
+        )
+
+    def __repr__(self) -> str:
+        pt = "pretrained" if self.pretrained else "scratch"
+        return f"RegNetY_3.2GF({pt}, in={self.in_shape}, out={self.out_size})"
+
+
+@register_model("regnet_y_8gf")
+class RegNetY8GF(RegNetBase):
+    """
+    RegNetY-8GF: High capacity for large-scale tasks.
+
+    ~37.4M backbone parameters, 8.0 GFLOPs. Use for maximum accuracy needs.
+
+    Recommended for:
+        - Very large datasets (>50k samples)
+        - Complex wave patterns
+        - HPC environments with ample GPU memory
+
+    Args:
+        in_shape: (H, W) image dimensions
+        out_size: Number of regression targets
+        pretrained: Use ImageNet pretrained weights (default: True)
+        dropout_rate: Dropout rate in head (default: 0.2)
+        freeze_backbone: Freeze backbone for fine-tuning (default: False)
+        regression_hidden: Hidden units in regression head (default: 256)
+
+    Example:
+        >>> model = RegNetY8GF(in_shape=(224, 224), out_size=3)
+        >>> x = torch.randn(4, 1, 224, 224)
+        >>> out = model(x)  # (4, 3)
+    """
+
+    def __init__(self, in_shape: tuple[int, int], out_size: int, **kwargs):
+        super().__init__(
+            in_shape=in_shape,
+            out_size=out_size,
+            model_fn=regnet_y_8gf,
+            weights_class=RegNet_Y_8GF_Weights,
+            **kwargs,
+        )
+
+    def __repr__(self) -> str:
+        pt = "pretrained" if self.pretrained else "scratch"
+        return f"RegNetY_8GF({pt}, in={self.in_shape}, out={self.out_size})"