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

wavedl/models/efficientnetv2.py

@@ -1,315 +1,315 @@
-"""
-EfficientNetV2: Faster Training and Better Accuracy
-====================================================
-
-Next-generation EfficientNet with improved training efficiency and performance.
-EfficientNetV2 replaces early depthwise convolutions with fused MBConv blocks,
-enabling 2-4× faster training while achieving better accuracy.
-
-**Key Improvements over EfficientNet**:
-    - Fused-MBConv in early stages (faster on accelerators)
-    - Progressive learning support (start small, grow)
-    - Better NAS-optimized architecture
-
-**Variants**:
-    - efficientnet_v2_s: Small (21.5M params) - Recommended default
-    - efficientnet_v2_m: Medium (54.1M params) - Higher accuracy
-    - efficientnet_v2_l: Large (118.5M params) - Maximum accuracy
-
-**Note**: EfficientNetV2 is 2D-only. For 1D data, use TCN. For 3D data, use ResNet3D.
-
-References:
-    Tan, M., & Le, Q. (2021). EfficientNetV2: Smaller Models and Faster Training.
-    ICML 2021. https://arxiv.org/abs/2104.00298
-
-Author: Ductho Le (ductho.le@outlook.com)
-"""
-
-from typing import Any
-
-import torch
-import torch.nn as nn
-
-
-try:
-    from torchvision.models import (
-        EfficientNet_V2_L_Weights,
-        EfficientNet_V2_M_Weights,
-        EfficientNet_V2_S_Weights,
-        efficientnet_v2_l,
-        efficientnet_v2_m,
-        efficientnet_v2_s,
-    )
-
-    EFFICIENTNETV2_AVAILABLE = True
-except ImportError:
-    EFFICIENTNETV2_AVAILABLE = False
-
-from wavedl.models.base import BaseModel
-from wavedl.models.registry import register_model
-
-
-class EfficientNetV2Base(BaseModel):
-    """
-    Base EfficientNetV2 class for regression tasks.
-
-    Wraps torchvision EfficientNetV2 with:
-    - Optional pretrained weights (ImageNet-1K)
-    - Automatic input channel adaptation (grayscale → 3ch)
-    - Custom multi-layer regression head
-
-    Compared to EfficientNet (V1):
-    - 2-4× faster training on GPU/TPU
-    - Better accuracy at similar parameter counts
-    - More efficient at higher resolutions
-
-    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 = 0.3,
-        freeze_backbone: bool = False,
-        regression_hidden: int = 512,
-        **kwargs,
-    ):
-        """
-        Initialize EfficientNetV2 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: 0.3)
-            freeze_backbone: Freeze backbone for fine-tuning (default: False)
-            regression_hidden: Hidden units in regression head (default: 512)
-        """
-        super().__init__(in_shape, out_size)
-
-        if not EFFICIENTNETV2_AVAILABLE:
-            raise ImportError(
-                "torchvision >= 0.13 is required for EfficientNetV2. "
-                "Install with: pip install torchvision>=0.13"
-            )
-
-        if len(in_shape) != 2:
-            raise ValueError(
-                f"EfficientNetV2 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)
-
-        # Get classifier input features (before the final classification layer)
-        in_features = self.backbone.classifier[1].in_features
-
-        # Replace classifier with regression head
-        # EfficientNetV2 benefits from a deeper regression head
-        self.backbone.classifier = nn.Sequential(
-            nn.Dropout(dropout_rate),
-            nn.Linear(in_features, regression_hidden),
-            nn.SiLU(inplace=True),  # SiLU (Swish) matches EfficientNet's activation
-            nn.Dropout(dropout_rate * 0.5),
-            nn.Linear(regression_hidden, regression_hidden // 2),
-            nn.SiLU(inplace=True),
-            nn.Linear(regression_hidden // 2, 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.features[0][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.features[0][0] = new_conv
-
-    def _freeze_backbone(self):
-        """Freeze all backbone parameters except the classifier."""
-        for name, param in self.backbone.named_parameters():
-            if "classifier" 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 EfficientNetV2."""
-        return {
-            "pretrained": True,
-            "dropout_rate": 0.3,
-            "freeze_backbone": False,
-            "regression_hidden": 512,
-        }
-
-
-# =============================================================================
-# REGISTERED MODEL VARIANTS
-# =============================================================================
-
-
-@register_model("efficientnet_v2_s")
-class EfficientNetV2S(EfficientNetV2Base):
-    """
-    EfficientNetV2-S: Small variant, recommended default.
-
-    ~21.5M parameters. Best balance of speed and accuracy for most tasks.
-    2× faster training than EfficientNet-B4 with better accuracy.
-
-    Recommended for:
-        - Default choice for 2D wave data
-        - Moderate compute budgets
-        - When training speed matters
-
-    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: 0.3)
-        freeze_backbone: Freeze backbone for fine-tuning (default: False)
-        regression_hidden: Hidden units in regression head (default: 512)
-
-    Example:
-        >>> model = EfficientNetV2S(in_shape=(500, 500), out_size=3)
-        >>> x = torch.randn(4, 1, 500, 500)
-        >>> 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=efficientnet_v2_s,
-            weights_class=EfficientNet_V2_S_Weights,
-            **kwargs,
-        )
-
-    def __repr__(self) -> str:
-        pt = "pretrained" if self.pretrained else "scratch"
-        return f"EfficientNetV2_S({pt}, in={self.in_shape}, out={self.out_size})"
-
-
-@register_model("efficientnet_v2_m")
-class EfficientNetV2M(EfficientNetV2Base):
-    """
-    EfficientNetV2-M: Medium variant for higher accuracy.
-
-    ~54.1M parameters. Use when accuracy is more important than speed.
-
-    Recommended for:
-        - Large datasets (>50k samples)
-        - Complex wave patterns
-        - When compute is not a bottleneck
-
-    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: 0.3)
-        freeze_backbone: Freeze backbone for fine-tuning (default: False)
-        regression_hidden: Hidden units in regression head (default: 512)
-
-    Example:
-        >>> model = EfficientNetV2M(in_shape=(500, 500), out_size=3)
-        >>> x = torch.randn(4, 1, 500, 500)
-        >>> 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=efficientnet_v2_m,
-            weights_class=EfficientNet_V2_M_Weights,
-            **kwargs,
-        )
-
-    def __repr__(self) -> str:
-        pt = "pretrained" if self.pretrained else "scratch"
-        return f"EfficientNetV2_M({pt}, in={self.in_shape}, out={self.out_size})"
-
-
-@register_model("efficientnet_v2_l")
-class EfficientNetV2L(EfficientNetV2Base):
-    """
-    EfficientNetV2-L: Large variant for maximum accuracy.
-
-    ~118.5M parameters. Use only with large datasets and sufficient compute.
-
-    Recommended for:
-        - Very large datasets (>100k samples)
-        - When maximum accuracy is critical
-        - 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: 0.3)
-        freeze_backbone: Freeze backbone for fine-tuning (default: False)
-        regression_hidden: Hidden units in regression head (default: 512)
-
-    Example:
-        >>> model = EfficientNetV2L(in_shape=(500, 500), out_size=3)
-        >>> x = torch.randn(4, 1, 500, 500)
-        >>> 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=efficientnet_v2_l,
-            weights_class=EfficientNet_V2_L_Weights,
-            **kwargs,
-        )
-
-    def __repr__(self) -> str:
-        pt = "pretrained" if self.pretrained else "scratch"
-        return f"EfficientNetV2_L({pt}, in={self.in_shape}, out={self.out_size})"
+"""
+EfficientNetV2: Faster Training and Better Accuracy
+====================================================
+
+Next-generation EfficientNet with improved training efficiency and performance.
+EfficientNetV2 replaces early depthwise convolutions with fused MBConv blocks,
+enabling 2-4× faster training while achieving better accuracy.
+
+**Key Improvements over EfficientNet**:
+    - Fused-MBConv in early stages (faster on accelerators)
+    - Progressive learning support (start small, grow)
+    - Better NAS-optimized architecture
+
+**Variants**:
+    - efficientnet_v2_s: Small (21.5M params) - Recommended default
+    - efficientnet_v2_m: Medium (54.1M params) - Higher accuracy
+    - efficientnet_v2_l: Large (118.5M params) - Maximum accuracy
+
+**Note**: EfficientNetV2 is 2D-only. For 1D data, use TCN. For 3D data, use ResNet3D.
+
+References:
+    Tan, M., & Le, Q. (2021). EfficientNetV2: Smaller Models and Faster Training.
+    ICML 2021. https://arxiv.org/abs/2104.00298
+
+Author: Ductho Le (ductho.le@outlook.com)
+"""
+
+from typing import Any
+
+import torch
+import torch.nn as nn
+
+
+try:
+    from torchvision.models import (
+        EfficientNet_V2_L_Weights,
+        EfficientNet_V2_M_Weights,
+        EfficientNet_V2_S_Weights,
+        efficientnet_v2_l,
+        efficientnet_v2_m,
+        efficientnet_v2_s,
+    )
+
+    EFFICIENTNETV2_AVAILABLE = True
+except ImportError:
+    EFFICIENTNETV2_AVAILABLE = False
+
+from wavedl.models.base import BaseModel
+from wavedl.models.registry import register_model
+
+
+class EfficientNetV2Base(BaseModel):
+    """
+    Base EfficientNetV2 class for regression tasks.
+
+    Wraps torchvision EfficientNetV2 with:
+    - Optional pretrained weights (ImageNet-1K)
+    - Automatic input channel adaptation (grayscale → 3ch)
+    - Custom multi-layer regression head
+
+    Compared to EfficientNet (V1):
+    - 2-4× faster training on GPU/TPU
+    - Better accuracy at similar parameter counts
+    - More efficient at higher resolutions
+
+    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 = 0.3,
+        freeze_backbone: bool = False,
+        regression_hidden: int = 512,
+        **kwargs,
+    ):
+        """
+        Initialize EfficientNetV2 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: 0.3)
+            freeze_backbone: Freeze backbone for fine-tuning (default: False)
+            regression_hidden: Hidden units in regression head (default: 512)
+        """
+        super().__init__(in_shape, out_size)
+
+        if not EFFICIENTNETV2_AVAILABLE:
+            raise ImportError(
+                "torchvision >= 0.13 is required for EfficientNetV2. "
+                "Install with: pip install torchvision>=0.13"
+            )
+
+        if len(in_shape) != 2:
+            raise ValueError(
+                f"EfficientNetV2 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)
+
+        # Get classifier input features (before the final classification layer)
+        in_features = self.backbone.classifier[1].in_features
+
+        # Replace classifier with regression head
+        # EfficientNetV2 benefits from a deeper regression head
+        self.backbone.classifier = nn.Sequential(
+            nn.Dropout(dropout_rate),
+            nn.Linear(in_features, regression_hidden),
+            nn.SiLU(inplace=True),  # SiLU (Swish) matches EfficientNet's activation
+            nn.Dropout(dropout_rate * 0.5),
+            nn.Linear(regression_hidden, regression_hidden // 2),
+            nn.SiLU(inplace=True),
+            nn.Linear(regression_hidden // 2, 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.features[0][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.features[0][0] = new_conv
+
+    def _freeze_backbone(self):
+        """Freeze all backbone parameters except the classifier."""
+        for name, param in self.backbone.named_parameters():
+            if "classifier" 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 EfficientNetV2."""
+        return {
+            "pretrained": True,
+            "dropout_rate": 0.3,
+            "freeze_backbone": False,
+            "regression_hidden": 512,
+        }
+
+
+# =============================================================================
+# REGISTERED MODEL VARIANTS
+# =============================================================================
+
+
+@register_model("efficientnet_v2_s")
+class EfficientNetV2S(EfficientNetV2Base):
+    """
+    EfficientNetV2-S: Small variant, recommended default.
+
+    ~20.2M backbone parameters. Best balance of speed and accuracy for most tasks.
+    2× faster training than EfficientNet-B4 with better accuracy.
+
+    Recommended for:
+        - Default choice for 2D wave data
+        - Moderate compute budgets
+        - When training speed matters
+
+    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: 0.3)
+        freeze_backbone: Freeze backbone for fine-tuning (default: False)
+        regression_hidden: Hidden units in regression head (default: 512)
+
+    Example:
+        >>> model = EfficientNetV2S(in_shape=(500, 500), out_size=3)
+        >>> x = torch.randn(4, 1, 500, 500)
+        >>> 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=efficientnet_v2_s,
+            weights_class=EfficientNet_V2_S_Weights,
+            **kwargs,
+        )
+
+    def __repr__(self) -> str:
+        pt = "pretrained" if self.pretrained else "scratch"
+        return f"EfficientNetV2_S({pt}, in={self.in_shape}, out={self.out_size})"
+
+
+@register_model("efficientnet_v2_m")
+class EfficientNetV2M(EfficientNetV2Base):
+    """
+    EfficientNetV2-M: Medium variant for higher accuracy.
+
+    ~52.9M backbone parameters. Use when accuracy is more important than speed.
+
+    Recommended for:
+        - Large datasets (>50k samples)
+        - Complex wave patterns
+        - When compute is not a bottleneck
+
+    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: 0.3)
+        freeze_backbone: Freeze backbone for fine-tuning (default: False)
+        regression_hidden: Hidden units in regression head (default: 512)
+
+    Example:
+        >>> model = EfficientNetV2M(in_shape=(500, 500), out_size=3)
+        >>> x = torch.randn(4, 1, 500, 500)
+        >>> 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=efficientnet_v2_m,
+            weights_class=EfficientNet_V2_M_Weights,
+            **kwargs,
+        )
+
+    def __repr__(self) -> str:
+        pt = "pretrained" if self.pretrained else "scratch"
+        return f"EfficientNetV2_M({pt}, in={self.in_shape}, out={self.out_size})"
+
+
+@register_model("efficientnet_v2_l")
+class EfficientNetV2L(EfficientNetV2Base):
+    """
+    EfficientNetV2-L: Large variant for maximum accuracy.
+
+    ~117.2M backbone parameters. Use only with large datasets and sufficient compute.
+
+    Recommended for:
+        - Very large datasets (>100k samples)
+        - When maximum accuracy is critical
+        - 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: 0.3)
+        freeze_backbone: Freeze backbone for fine-tuning (default: False)
+        regression_hidden: Hidden units in regression head (default: 512)
+
+    Example:
+        >>> model = EfficientNetV2L(in_shape=(500, 500), out_size=3)
+        >>> x = torch.randn(4, 1, 500, 500)
+        >>> 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=efficientnet_v2_l,
+            weights_class=EfficientNet_V2_L_Weights,
+            **kwargs,
+        )
+
+    def __repr__(self) -> str:
+        pt = "pretrained" if self.pretrained else "scratch"
+        return f"EfficientNetV2_L({pt}, in={self.in_shape}, out={self.out_size})"