सामग्री पर जाएं

के लिए संदर्भ ultralytics/models/sam/modules/tiny_encoder.py

नोट

यह फ़ाइल यहाँ उपलब्ध है https://github.com/ultralytics/ultralytics/बूँद/मुख्य/ultralytics/मॉडल/sam/मॉड्यूल/tiny_encoder.py का उपयोग करें। यदि आप कोई समस्या देखते हैं तो कृपया पुल अनुरोध का योगदान करके इसे ठीक करने में मदद करें 🛠️। 🙏 धन्यवाद !



ultralytics.models.sam.modules.tiny_encoder.Conv2d_BN

का रूप: Sequential

एक अनुक्रमिक कंटेनर जो बैच सामान्यीकरण के बाद 2D कनवल्शन करता है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
class Conv2d_BN(torch.nn.Sequential):
    """A sequential container that performs 2D convolution followed by batch normalization."""

    def __init__(self, a, b, ks=1, stride=1, pad=0, dilation=1, groups=1, bn_weight_init=1):
        """Initializes the MBConv model with given input channels, output channels, expansion ratio, activation, and
        drop path.
        """
        super().__init__()
        self.add_module("c", torch.nn.Conv2d(a, b, ks, stride, pad, dilation, groups, bias=False))
        bn = torch.nn.BatchNorm2d(b)
        torch.nn.init.constant_(bn.weight, bn_weight_init)
        torch.nn.init.constant_(bn.bias, 0)
        self.add_module("bn", bn)

__init__(a, b, ks=1, stride=1, pad=0, dilation=1, groups=1, bn_weight_init=1)

दिए गए इनपुट चैनलों, आउटपुट चैनलों, विस्तार अनुपात, सक्रियण और के साथ MBConv मॉडल को इनिशियलाइज़ करता है ड्रॉप पाथ।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def __init__(self, a, b, ks=1, stride=1, pad=0, dilation=1, groups=1, bn_weight_init=1):
    """Initializes the MBConv model with given input channels, output channels, expansion ratio, activation, and
    drop path.
    """
    super().__init__()
    self.add_module("c", torch.nn.Conv2d(a, b, ks, stride, pad, dilation, groups, bias=False))
    bn = torch.nn.BatchNorm2d(b)
    torch.nn.init.constant_(bn.weight, bn_weight_init)
    torch.nn.init.constant_(bn.bias, 0)
    self.add_module("bn", bn)



ultralytics.models.sam.modules.tiny_encoder.PatchEmbed

का रूप: Module

छवियों को पैच में एम्बेड करता है और उन्हें एक निर्दिष्ट एम्बेडिंग आयाम में प्रोजेक्ट करता है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
class PatchEmbed(nn.Module):
    """Embeds images into patches and projects them into a specified embedding dimension."""

    def __init__(self, in_chans, embed_dim, resolution, activation):
        """Initialize the PatchMerging class with specified input, output dimensions, resolution and activation
        function.
        """
        super().__init__()
        img_size: Tuple[int, int] = to_2tuple(resolution)
        self.patches_resolution = (img_size[0] // 4, img_size[1] // 4)
        self.num_patches = self.patches_resolution[0] * self.patches_resolution[1]
        self.in_chans = in_chans
        self.embed_dim = embed_dim
        n = embed_dim
        self.seq = nn.Sequential(
            Conv2d_BN(in_chans, n // 2, 3, 2, 1),
            activation(),
            Conv2d_BN(n // 2, n, 3, 2, 1),
        )

    def forward(self, x):
        """Runs input tensor 'x' through the PatchMerging model's sequence of operations."""
        return self.seq(x)

__init__(in_chans, embed_dim, resolution, activation)

निर्दिष्ट इनपुट, आउटपुट आयाम, रिज़ॉल्यूशन और सक्रियण के साथ पैचमर्जिंग क्लास को इनिशियलाइज़ करें फलन।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def __init__(self, in_chans, embed_dim, resolution, activation):
    """Initialize the PatchMerging class with specified input, output dimensions, resolution and activation
    function.
    """
    super().__init__()
    img_size: Tuple[int, int] = to_2tuple(resolution)
    self.patches_resolution = (img_size[0] // 4, img_size[1] // 4)
    self.num_patches = self.patches_resolution[0] * self.patches_resolution[1]
    self.in_chans = in_chans
    self.embed_dim = embed_dim
    n = embed_dim
    self.seq = nn.Sequential(
        Conv2d_BN(in_chans, n // 2, 3, 2, 1),
        activation(),
        Conv2d_BN(n // 2, n, 3, 2, 1),
    )

forward(x)

इनपुट चलाता है tensor पैचमर्जिंग मॉडल के संचालन के अनुक्रम के माध्यम से 'x'।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def forward(self, x):
    """Runs input tensor 'x' through the PatchMerging model's sequence of operations."""
    return self.seq(x)



ultralytics.models.sam.modules.tiny_encoder.MBConv

का रूप: Module

मोबाइल इनवर्टेड बॉटलनेक कॉन्व (MBConv) परत, EfficientNet आर्किटेक्चर का हिस्सा।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
class MBConv(nn.Module):
    """Mobile Inverted Bottleneck Conv (MBConv) layer, part of the EfficientNet architecture."""

    def __init__(self, in_chans, out_chans, expand_ratio, activation, drop_path):
        """Initializes a convolutional layer with specified dimensions, input resolution, depth, and activation
        function.
        """
        super().__init__()
        self.in_chans = in_chans
        self.hidden_chans = int(in_chans * expand_ratio)
        self.out_chans = out_chans

        self.conv1 = Conv2d_BN(in_chans, self.hidden_chans, ks=1)
        self.act1 = activation()

        self.conv2 = Conv2d_BN(self.hidden_chans, self.hidden_chans, ks=3, stride=1, pad=1, groups=self.hidden_chans)
        self.act2 = activation()

        self.conv3 = Conv2d_BN(self.hidden_chans, out_chans, ks=1, bn_weight_init=0.0)
        self.act3 = activation()

        # NOTE: `DropPath` is needed only for training.
        # self.drop_path = DropPath(drop_path) if drop_path > 0. else nn.Identity()
        self.drop_path = nn.Identity()

    def forward(self, x):
        """Implements the forward pass for the model architecture."""
        shortcut = x
        x = self.conv1(x)
        x = self.act1(x)
        x = self.conv2(x)
        x = self.act2(x)
        x = self.conv3(x)
        x = self.drop_path(x)
        x += shortcut
        return self.act3(x)

__init__(in_chans, out_chans, expand_ratio, activation, drop_path)

निर्दिष्ट आयामों, इनपुट रिज़ॉल्यूशन, गहराई और सक्रियण के साथ एक दृढ़ परत को इनिशियलाइज़ करता है फलन।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def __init__(self, in_chans, out_chans, expand_ratio, activation, drop_path):
    """Initializes a convolutional layer with specified dimensions, input resolution, depth, and activation
    function.
    """
    super().__init__()
    self.in_chans = in_chans
    self.hidden_chans = int(in_chans * expand_ratio)
    self.out_chans = out_chans

    self.conv1 = Conv2d_BN(in_chans, self.hidden_chans, ks=1)
    self.act1 = activation()

    self.conv2 = Conv2d_BN(self.hidden_chans, self.hidden_chans, ks=3, stride=1, pad=1, groups=self.hidden_chans)
    self.act2 = activation()

    self.conv3 = Conv2d_BN(self.hidden_chans, out_chans, ks=1, bn_weight_init=0.0)
    self.act3 = activation()

    # NOTE: `DropPath` is needed only for training.
    # self.drop_path = DropPath(drop_path) if drop_path > 0. else nn.Identity()
    self.drop_path = nn.Identity()

forward(x)

मॉडल आर्किटेक्चर के लिए फॉरवर्ड पास को लागू करता है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def forward(self, x):
    """Implements the forward pass for the model architecture."""
    shortcut = x
    x = self.conv1(x)
    x = self.act1(x)
    x = self.conv2(x)
    x = self.act2(x)
    x = self.conv3(x)
    x = self.drop_path(x)
    x += shortcut
    return self.act3(x)



ultralytics.models.sam.modules.tiny_encoder.PatchMerging

का रूप: Module

सुविधा मानचित्र में पड़ोसी पैच को मर्ज करता है और एक नए आयाम में प्रोजेक्ट करता है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
class PatchMerging(nn.Module):
    """Merges neighboring patches in the feature map and projects to a new dimension."""

    def __init__(self, input_resolution, dim, out_dim, activation):
        """Initializes the ConvLayer with specific dimension, input resolution, depth, activation, drop path, and other
        optional parameters.
        """
        super().__init__()

        self.input_resolution = input_resolution
        self.dim = dim
        self.out_dim = out_dim
        self.act = activation()
        self.conv1 = Conv2d_BN(dim, out_dim, 1, 1, 0)
        stride_c = 1 if out_dim in {320, 448, 576} else 2
        self.conv2 = Conv2d_BN(out_dim, out_dim, 3, stride_c, 1, groups=out_dim)
        self.conv3 = Conv2d_BN(out_dim, out_dim, 1, 1, 0)

    def forward(self, x):
        """Applies forward pass on the input utilizing convolution and activation layers, and returns the result."""
        if x.ndim == 3:
            H, W = self.input_resolution
            B = len(x)
            # (B, C, H, W)
            x = x.view(B, H, W, -1).permute(0, 3, 1, 2)

        x = self.conv1(x)
        x = self.act(x)

        x = self.conv2(x)
        x = self.act(x)
        x = self.conv3(x)
        return x.flatten(2).transpose(1, 2)

__init__(input_resolution, dim, out_dim, activation)

विशिष्ट आयाम, इनपुट रिज़ॉल्यूशन, गहराई, सक्रियण, ड्रॉप पथ और अन्य के साथ ConvLayer को इनिशियलाइज़ करता है वैकल्पिक पैरामीटर।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def __init__(self, input_resolution, dim, out_dim, activation):
    """Initializes the ConvLayer with specific dimension, input resolution, depth, activation, drop path, and other
    optional parameters.
    """
    super().__init__()

    self.input_resolution = input_resolution
    self.dim = dim
    self.out_dim = out_dim
    self.act = activation()
    self.conv1 = Conv2d_BN(dim, out_dim, 1, 1, 0)
    stride_c = 1 if out_dim in {320, 448, 576} else 2
    self.conv2 = Conv2d_BN(out_dim, out_dim, 3, stride_c, 1, groups=out_dim)
    self.conv3 = Conv2d_BN(out_dim, out_dim, 1, 1, 0)

forward(x)

कनवल्शन और सक्रियण परतों का उपयोग करके इनपुट पर फॉरवर्ड पास लागू करता है, और परिणाम देता है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def forward(self, x):
    """Applies forward pass on the input utilizing convolution and activation layers, and returns the result."""
    if x.ndim == 3:
        H, W = self.input_resolution
        B = len(x)
        # (B, C, H, W)
        x = x.view(B, H, W, -1).permute(0, 3, 1, 2)

    x = self.conv1(x)
    x = self.act(x)

    x = self.conv2(x)
    x = self.act(x)
    x = self.conv3(x)
    return x.flatten(2).transpose(1, 2)



ultralytics.models.sam.modules.tiny_encoder.ConvLayer

का रूप: Module

कई MobileNetV3-शैली उल्टे अड़चन convolutions (MBConv) की विशेषता वाली दृढ़ परत।

वैकल्पिक रूप से आउटपुट पर डाउनसैंपल ऑपरेशन लागू करता है, और ग्रेडिएंट चेकपॉइंटिंग के लिए समर्थन प्रदान करता है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
class ConvLayer(nn.Module):
    """
    Convolutional Layer featuring multiple MobileNetV3-style inverted bottleneck convolutions (MBConv).

    Optionally applies downsample operations to the output, and provides support for gradient checkpointing.
    """

    def __init__(
        self,
        dim,
        input_resolution,
        depth,
        activation,
        drop_path=0.0,
        downsample=None,
        use_checkpoint=False,
        out_dim=None,
        conv_expand_ratio=4.0,
    ):
        """
        Initializes the ConvLayer with the given dimensions and settings.

        Args:
            dim (int): The dimensionality of the input and output.
            input_resolution (Tuple[int, int]): The resolution of the input image.
            depth (int): The number of MBConv layers in the block.
            activation (Callable): Activation function applied after each convolution.
            drop_path (Union[float, List[float]]): Drop path rate. Single float or a list of floats for each MBConv.
            downsample (Optional[Callable]): Function for downsampling the output. None to skip downsampling.
            use_checkpoint (bool): Whether to use gradient checkpointing to save memory.
            out_dim (Optional[int]): The dimensionality of the output. None means it will be the same as `dim`.
            conv_expand_ratio (float): Expansion ratio for the MBConv layers.
        """
        super().__init__()
        self.dim = dim
        self.input_resolution = input_resolution
        self.depth = depth
        self.use_checkpoint = use_checkpoint

        # Build blocks
        self.blocks = nn.ModuleList(
            [
                MBConv(
                    dim,
                    dim,
                    conv_expand_ratio,
                    activation,
                    drop_path[i] if isinstance(drop_path, list) else drop_path,
                )
                for i in range(depth)
            ]
        )

        # Patch merging layer
        self.downsample = (
            None
            if downsample is None
            else downsample(input_resolution, dim=dim, out_dim=out_dim, activation=activation)
        )

    def forward(self, x):
        """Processes the input through a series of convolutional layers and returns the activated output."""
        for blk in self.blocks:
            x = checkpoint.checkpoint(blk, x) if self.use_checkpoint else blk(x)
        return x if self.downsample is None else self.downsample(x)

__init__(dim, input_resolution, depth, activation, drop_path=0.0, downsample=None, use_checkpoint=False, out_dim=None, conv_expand_ratio=4.0)

दिए गए आयामों और सेटिंग्स के साथ ConvLayer को इनविजियलाइज़ करता है।

पैरामीटर:

नाम प्रकार या क़िस्‍म चूक
dim int

इनपुट और आउटपुट की आयामीता।

आवश्यक
input_resolution Tuple[int, int]

इनपुट छवि का संकल्प।

आवश्यक
depth int

ब्लॉक में MBConv परतों की संख्या।

आवश्यक
activation Callable

सक्रियण फ़ंक्शन प्रत्येक कनवल्शन के बाद लागू होता है।

आवश्यक
drop_path Union[float, List[float]]

ड्रॉप पथ दर। सिंगल फ्लोट या प्रत्येक MBConv के लिए फ्लोट्स की एक सूची।

0.0
downsample Optional[Callable]

आउटपुट को डाउनसैंपलिंग के लिए फ़ंक्शन। डाउनसैंपलिंग को छोड़ने के लिए कोई नहीं।

None
use_checkpoint bool

स्मृति को बचाने के लिए ढाल चेकपॉइंटिंग का उपयोग करना है या नहीं।

False
out_dim Optional[int]

आउटपुट की आयामीता। किसी का मतलब नहीं है कि यह वैसा ही होगा जैसा dim.

None
conv_expand_ratio float

MBConv परतों के लिए विस्तार अनुपात।

4.0
में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def __init__(
    self,
    dim,
    input_resolution,
    depth,
    activation,
    drop_path=0.0,
    downsample=None,
    use_checkpoint=False,
    out_dim=None,
    conv_expand_ratio=4.0,
):
    """
    Initializes the ConvLayer with the given dimensions and settings.

    Args:
        dim (int): The dimensionality of the input and output.
        input_resolution (Tuple[int, int]): The resolution of the input image.
        depth (int): The number of MBConv layers in the block.
        activation (Callable): Activation function applied after each convolution.
        drop_path (Union[float, List[float]]): Drop path rate. Single float or a list of floats for each MBConv.
        downsample (Optional[Callable]): Function for downsampling the output. None to skip downsampling.
        use_checkpoint (bool): Whether to use gradient checkpointing to save memory.
        out_dim (Optional[int]): The dimensionality of the output. None means it will be the same as `dim`.
        conv_expand_ratio (float): Expansion ratio for the MBConv layers.
    """
    super().__init__()
    self.dim = dim
    self.input_resolution = input_resolution
    self.depth = depth
    self.use_checkpoint = use_checkpoint

    # Build blocks
    self.blocks = nn.ModuleList(
        [
            MBConv(
                dim,
                dim,
                conv_expand_ratio,
                activation,
                drop_path[i] if isinstance(drop_path, list) else drop_path,
            )
            for i in range(depth)
        ]
    )

    # Patch merging layer
    self.downsample = (
        None
        if downsample is None
        else downsample(input_resolution, dim=dim, out_dim=out_dim, activation=activation)
    )

forward(x)

दृढ़ परतों की एक श्रृंखला के माध्यम से इनपुट को संसाधित करता है और सक्रिय आउटपुट लौटाता है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def forward(self, x):
    """Processes the input through a series of convolutional layers and returns the activated output."""
    for blk in self.blocks:
        x = checkpoint.checkpoint(blk, x) if self.use_checkpoint else blk(x)
    return x if self.downsample is None else self.downsample(x)



ultralytics.models.sam.modules.tiny_encoder.Mlp

का रूप: Module

ट्रांसफार्मर आर्किटेक्चर के लिए मल्टी-लेयर परसेप्ट्रॉन (एमएलपी)।

यह परत in_features के साथ एक इनपुट लेती है, परत सामान्यीकरण और दो पूरी तरह से जुड़ी परतों को लागू करती है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
class Mlp(nn.Module):
    """
    Multi-layer Perceptron (MLP) for transformer architectures.

    This layer takes an input with in_features, applies layer normalization and two fully-connected layers.
    """

    def __init__(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.0):
        """Initializes Attention module with the given parameters including dimension, key_dim, number of heads, etc."""
        super().__init__()
        out_features = out_features or in_features
        hidden_features = hidden_features or in_features
        self.norm = nn.LayerNorm(in_features)
        self.fc1 = nn.Linear(in_features, hidden_features)
        self.fc2 = nn.Linear(hidden_features, out_features)
        self.act = act_layer()
        self.drop = nn.Dropout(drop)

    def forward(self, x):
        """Applies operations on input x and returns modified x, runs downsample if not None."""
        x = self.norm(x)
        x = self.fc1(x)
        x = self.act(x)
        x = self.drop(x)
        x = self.fc2(x)
        return self.drop(x)

__init__(in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.0)

आयाम, key_dim, सिर की संख्या आदि सहित दिए गए मापदंडों के साथ ध्यान मॉड्यूल को प्रारंभ करता है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def __init__(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.0):
    """Initializes Attention module with the given parameters including dimension, key_dim, number of heads, etc."""
    super().__init__()
    out_features = out_features or in_features
    hidden_features = hidden_features or in_features
    self.norm = nn.LayerNorm(in_features)
    self.fc1 = nn.Linear(in_features, hidden_features)
    self.fc2 = nn.Linear(hidden_features, out_features)
    self.act = act_layer()
    self.drop = nn.Dropout(drop)

forward(x)

इनपुट x पर संचालन लागू करता है और संशोधित x लौटाता है, यदि कोई नहीं तो downsample चलाता है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def forward(self, x):
    """Applies operations on input x and returns modified x, runs downsample if not None."""
    x = self.norm(x)
    x = self.fc1(x)
    x = self.act(x)
    x = self.drop(x)
    x = self.fc2(x)
    return self.drop(x)



ultralytics.models.sam.modules.tiny_encoder.Attention

का रूप: Module

स्थानिक जागरूकता के लिए समर्थन के साथ बहु-सिर ध्यान मॉड्यूल, स्थानिक के आधार पर ध्यान पूर्वाग्रहों को लागू करना प्रस्ताव। संकल्प में स्थानिक पदों के बीच प्रत्येक अद्वितीय ऑफसेट के लिए प्रशिक्षित ध्यान पूर्वाग्रहों को लागू करता है ग्रिड।

विशेषताएँ:

नाम प्रकार या क़िस्‍म
ab Tensor

अनुमान के लिए कैश्ड ध्यान पूर्वाग्रह, प्रशिक्षण के दौरान हटा दिया गया।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
class Attention(torch.nn.Module):
    """
    Multi-head attention module with support for spatial awareness, applying attention biases based on spatial
    resolution. Implements trainable attention biases for each unique offset between spatial positions in the resolution
    grid.

    Attributes:
        ab (Tensor, optional): Cached attention biases for inference, deleted during training.
    """

    def __init__(
        self,
        dim,
        key_dim,
        num_heads=8,
        attn_ratio=4,
        resolution=(14, 14),
    ):
        """
        Initializes the Attention module.

        Args:
            dim (int): The dimensionality of the input and output.
            key_dim (int): The dimensionality of the keys and queries.
            num_heads (int, optional): Number of attention heads. Default is 8.
            attn_ratio (float, optional): Attention ratio, affecting the dimensions of the value vectors. Default is 4.
            resolution (Tuple[int, int], optional): Spatial resolution of the input feature map. Default is (14, 14).

        Raises:
            AssertionError: If `resolution` is not a tuple of length 2.
        """
        super().__init__()

        assert isinstance(resolution, tuple) and len(resolution) == 2
        self.num_heads = num_heads
        self.scale = key_dim**-0.5
        self.key_dim = key_dim
        self.nh_kd = nh_kd = key_dim * num_heads
        self.d = int(attn_ratio * key_dim)
        self.dh = int(attn_ratio * key_dim) * num_heads
        self.attn_ratio = attn_ratio
        h = self.dh + nh_kd * 2

        self.norm = nn.LayerNorm(dim)
        self.qkv = nn.Linear(dim, h)
        self.proj = nn.Linear(self.dh, dim)

        points = list(itertools.product(range(resolution[0]), range(resolution[1])))
        N = len(points)
        attention_offsets = {}
        idxs = []
        for p1 in points:
            for p2 in points:
                offset = (abs(p1[0] - p2[0]), abs(p1[1] - p2[1]))
                if offset not in attention_offsets:
                    attention_offsets[offset] = len(attention_offsets)
                idxs.append(attention_offsets[offset])
        self.attention_biases = torch.nn.Parameter(torch.zeros(num_heads, len(attention_offsets)))
        self.register_buffer("attention_bias_idxs", torch.LongTensor(idxs).view(N, N), persistent=False)

    @torch.no_grad()
    def train(self, mode=True):
        """Sets the module in training mode and handles attribute 'ab' based on the mode."""
        super().train(mode)
        if mode and hasattr(self, "ab"):
            del self.ab
        else:
            self.ab = self.attention_biases[:, self.attention_bias_idxs]

    def forward(self, x):  # x
        """Performs forward pass over the input tensor 'x' by applying normalization and querying keys/values."""
        B, N, _ = x.shape  # B, N, C

        # Normalization
        x = self.norm(x)

        qkv = self.qkv(x)
        # (B, N, num_heads, d)
        q, k, v = qkv.view(B, N, self.num_heads, -1).split([self.key_dim, self.key_dim, self.d], dim=3)
        # (B, num_heads, N, d)
        q = q.permute(0, 2, 1, 3)
        k = k.permute(0, 2, 1, 3)
        v = v.permute(0, 2, 1, 3)
        self.ab = self.ab.to(self.attention_biases.device)

        attn = (q @ k.transpose(-2, -1)) * self.scale + (
            self.attention_biases[:, self.attention_bias_idxs] if self.training else self.ab
        )
        attn = attn.softmax(dim=-1)
        x = (attn @ v).transpose(1, 2).reshape(B, N, self.dh)
        return self.proj(x)

__init__(dim, key_dim, num_heads=8, attn_ratio=4, resolution=(14, 14))

ध्यान मॉड्यूल को प्रारंभ करता है।

पैरामीटर:

नाम प्रकार या क़िस्‍म चूक
dim int

इनपुट और आउटपुट की आयामीता।

आवश्यक
key_dim int

कुंजियों और प्रश्नों की आयामीता।

आवश्यक
num_heads int

ध्यान प्रमुखों की संख्या। डिफ़ॉल्ट 8 है।

8
attn_ratio float

ध्यान अनुपात, मूल्य वैक्टर के आयामों को प्रभावित करता है। डिफ़ॉल्ट 4 है।

4
resolution Tuple[int, int]

इनपुट सुविधा मानचित्र का स्थानिक रिज़ॉल्यूशन। डिफ़ॉल्ट (14, 14) है।

(14, 14)

उठाती:

प्रकार या क़िस्‍म
AssertionError

अगर resolution लंबाई 2 का टपल नहीं है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def __init__(
    self,
    dim,
    key_dim,
    num_heads=8,
    attn_ratio=4,
    resolution=(14, 14),
):
    """
    Initializes the Attention module.

    Args:
        dim (int): The dimensionality of the input and output.
        key_dim (int): The dimensionality of the keys and queries.
        num_heads (int, optional): Number of attention heads. Default is 8.
        attn_ratio (float, optional): Attention ratio, affecting the dimensions of the value vectors. Default is 4.
        resolution (Tuple[int, int], optional): Spatial resolution of the input feature map. Default is (14, 14).

    Raises:
        AssertionError: If `resolution` is not a tuple of length 2.
    """
    super().__init__()

    assert isinstance(resolution, tuple) and len(resolution) == 2
    self.num_heads = num_heads
    self.scale = key_dim**-0.5
    self.key_dim = key_dim
    self.nh_kd = nh_kd = key_dim * num_heads
    self.d = int(attn_ratio * key_dim)
    self.dh = int(attn_ratio * key_dim) * num_heads
    self.attn_ratio = attn_ratio
    h = self.dh + nh_kd * 2

    self.norm = nn.LayerNorm(dim)
    self.qkv = nn.Linear(dim, h)
    self.proj = nn.Linear(self.dh, dim)

    points = list(itertools.product(range(resolution[0]), range(resolution[1])))
    N = len(points)
    attention_offsets = {}
    idxs = []
    for p1 in points:
        for p2 in points:
            offset = (abs(p1[0] - p2[0]), abs(p1[1] - p2[1]))
            if offset not in attention_offsets:
                attention_offsets[offset] = len(attention_offsets)
            idxs.append(attention_offsets[offset])
    self.attention_biases = torch.nn.Parameter(torch.zeros(num_heads, len(attention_offsets)))
    self.register_buffer("attention_bias_idxs", torch.LongTensor(idxs).view(N, N), persistent=False)

forward(x)

इनपुट पर फॉरवर्ड पास करता है tensor सामान्यीकरण लागू करके और कुंजियों/मानों को क्वेरी करके 'x'।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def forward(self, x):  # x
    """Performs forward pass over the input tensor 'x' by applying normalization and querying keys/values."""
    B, N, _ = x.shape  # B, N, C

    # Normalization
    x = self.norm(x)

    qkv = self.qkv(x)
    # (B, N, num_heads, d)
    q, k, v = qkv.view(B, N, self.num_heads, -1).split([self.key_dim, self.key_dim, self.d], dim=3)
    # (B, num_heads, N, d)
    q = q.permute(0, 2, 1, 3)
    k = k.permute(0, 2, 1, 3)
    v = v.permute(0, 2, 1, 3)
    self.ab = self.ab.to(self.attention_biases.device)

    attn = (q @ k.transpose(-2, -1)) * self.scale + (
        self.attention_biases[:, self.attention_bias_idxs] if self.training else self.ab
    )
    attn = attn.softmax(dim=-1)
    x = (attn @ v).transpose(1, 2).reshape(B, N, self.dh)
    return self.proj(x)

train(mode=True)

मॉड्यूल को प्रशिक्षण मोड में सेट करता है और मोड के आधार पर विशेषता 'ab' को संभालता है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
@torch.no_grad()
def train(self, mode=True):
    """Sets the module in training mode and handles attribute 'ab' based on the mode."""
    super().train(mode)
    if mode and hasattr(self, "ab"):
        del self.ab
    else:
        self.ab = self.attention_biases[:, self.attention_bias_idxs]



ultralytics.models.sam.modules.tiny_encoder.TinyViTBlock

का रूप: Module

TinyViT ब्लॉक जो इनपुट पर आत्म-ध्यान और स्थानीय कनवल्शन लागू करता है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
class TinyViTBlock(nn.Module):
    """TinyViT Block that applies self-attention and a local convolution to the input."""

    def __init__(
        self,
        dim,
        input_resolution,
        num_heads,
        window_size=7,
        mlp_ratio=4.0,
        drop=0.0,
        drop_path=0.0,
        local_conv_size=3,
        activation=nn.GELU,
    ):
        """
        Initializes the TinyViTBlock.

        Args:
            dim (int): The dimensionality of the input and output.
            input_resolution (Tuple[int, int]): Spatial resolution of the input feature map.
            num_heads (int): Number of attention heads.
            window_size (int, optional): Window size for attention. Default is 7.
            mlp_ratio (float, optional): Ratio of mlp hidden dim to embedding dim. Default is 4.
            drop (float, optional): Dropout rate. Default is 0.
            drop_path (float, optional): Stochastic depth rate. Default is 0.
            local_conv_size (int, optional): The kernel size of the local convolution. Default is 3.
            activation (torch.nn, optional): Activation function for MLP. Default is nn.GELU.

        Raises:
            AssertionError: If `window_size` is not greater than 0.
            AssertionError: If `dim` is not divisible by `num_heads`.
        """
        super().__init__()
        self.dim = dim
        self.input_resolution = input_resolution
        self.num_heads = num_heads
        assert window_size > 0, "window_size must be greater than 0"
        self.window_size = window_size
        self.mlp_ratio = mlp_ratio

        # NOTE: `DropPath` is needed only for training.
        # self.drop_path = DropPath(drop_path) if drop_path > 0. else nn.Identity()
        self.drop_path = nn.Identity()

        assert dim % num_heads == 0, "dim must be divisible by num_heads"
        head_dim = dim // num_heads

        window_resolution = (window_size, window_size)
        self.attn = Attention(dim, head_dim, num_heads, attn_ratio=1, resolution=window_resolution)

        mlp_hidden_dim = int(dim * mlp_ratio)
        mlp_activation = activation
        self.mlp = Mlp(in_features=dim, hidden_features=mlp_hidden_dim, act_layer=mlp_activation, drop=drop)

        pad = local_conv_size // 2
        self.local_conv = Conv2d_BN(dim, dim, ks=local_conv_size, stride=1, pad=pad, groups=dim)

    def forward(self, x):
        """Applies attention-based transformation or padding to input 'x' before passing it through a local
        convolution.
        """
        h, w = self.input_resolution
        b, l, c = x.shape
        assert l == h * w, "input feature has wrong size"
        res_x = x
        if h == self.window_size and w == self.window_size:
            x = self.attn(x)
        else:
            x = x.view(b, h, w, c)
            pad_b = (self.window_size - h % self.window_size) % self.window_size
            pad_r = (self.window_size - w % self.window_size) % self.window_size
            padding = pad_b > 0 or pad_r > 0

            if padding:
                x = F.pad(x, (0, 0, 0, pad_r, 0, pad_b))

            pH, pW = h + pad_b, w + pad_r
            nH = pH // self.window_size
            nW = pW // self.window_size
            # Window partition
            x = (
                x.view(b, nH, self.window_size, nW, self.window_size, c)
                .transpose(2, 3)
                .reshape(b * nH * nW, self.window_size * self.window_size, c)
            )
            x = self.attn(x)
            # Window reverse
            x = x.view(b, nH, nW, self.window_size, self.window_size, c).transpose(2, 3).reshape(b, pH, pW, c)

            if padding:
                x = x[:, :h, :w].contiguous()

            x = x.view(b, l, c)

        x = res_x + self.drop_path(x)

        x = x.transpose(1, 2).reshape(b, c, h, w)
        x = self.local_conv(x)
        x = x.view(b, c, l).transpose(1, 2)

        return x + self.drop_path(self.mlp(x))

    def extra_repr(self) -> str:
        """Returns a formatted string representing the TinyViTBlock's parameters: dimension, input resolution, number of
        attentions heads, window size, and MLP ratio.
        """
        return (
            f"dim={self.dim}, input_resolution={self.input_resolution}, num_heads={self.num_heads}, "
            f"window_size={self.window_size}, mlp_ratio={self.mlp_ratio}"
        )

__init__(dim, input_resolution, num_heads, window_size=7, mlp_ratio=4.0, drop=0.0, drop_path=0.0, local_conv_size=3, activation=nn.GELU)

TinyViTBlock को इनिशियलाइज़ करता है।

पैरामीटर:

नाम प्रकार या क़िस्‍म चूक
dim int

इनपुट और आउटपुट की आयामीता।

आवश्यक
input_resolution Tuple[int, int]

इनपुट सुविधा मानचित्र का स्थानिक रिज़ॉल्यूशन।

आवश्यक
num_heads int

ध्यान प्रमुखों की संख्या।

आवश्यक
window_size int

ध्यान के लिए खिड़की का आकार। डिफ़ॉल्ट 7 है।

7
mlp_ratio float

एमएलपी का अनुपात मंद को एम्बेडिंग मंद करने के लिए छिपा हुआ है। डिफ़ॉल्ट 4 है।

4.0
drop float

ड्रॉपआउट दर। डिफ़ॉल्ट 0 है।

0.0
drop_path float

स्टोकेस्टिक गहराई दर। डिफ़ॉल्ट 0 है।

0.0
local_conv_size int

स्थानीय कनवल्शन का कर्नेल आकार। डिफ़ॉल्ट 3 है।

3
activation nn

एमएलपी के लिए सक्रियण समारोह। डिफ़ॉल्ट nn है. गेलु।

GELU

उठाती:

प्रकार या क़िस्‍म
AssertionError

अगर window_size 0 से बड़ा नहीं है।

AssertionError

अगर dim द्वारा विभाज्य नहीं है num_heads.

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def __init__(
    self,
    dim,
    input_resolution,
    num_heads,
    window_size=7,
    mlp_ratio=4.0,
    drop=0.0,
    drop_path=0.0,
    local_conv_size=3,
    activation=nn.GELU,
):
    """
    Initializes the TinyViTBlock.

    Args:
        dim (int): The dimensionality of the input and output.
        input_resolution (Tuple[int, int]): Spatial resolution of the input feature map.
        num_heads (int): Number of attention heads.
        window_size (int, optional): Window size for attention. Default is 7.
        mlp_ratio (float, optional): Ratio of mlp hidden dim to embedding dim. Default is 4.
        drop (float, optional): Dropout rate. Default is 0.
        drop_path (float, optional): Stochastic depth rate. Default is 0.
        local_conv_size (int, optional): The kernel size of the local convolution. Default is 3.
        activation (torch.nn, optional): Activation function for MLP. Default is nn.GELU.

    Raises:
        AssertionError: If `window_size` is not greater than 0.
        AssertionError: If `dim` is not divisible by `num_heads`.
    """
    super().__init__()
    self.dim = dim
    self.input_resolution = input_resolution
    self.num_heads = num_heads
    assert window_size > 0, "window_size must be greater than 0"
    self.window_size = window_size
    self.mlp_ratio = mlp_ratio

    # NOTE: `DropPath` is needed only for training.
    # self.drop_path = DropPath(drop_path) if drop_path > 0. else nn.Identity()
    self.drop_path = nn.Identity()

    assert dim % num_heads == 0, "dim must be divisible by num_heads"
    head_dim = dim // num_heads

    window_resolution = (window_size, window_size)
    self.attn = Attention(dim, head_dim, num_heads, attn_ratio=1, resolution=window_resolution)

    mlp_hidden_dim = int(dim * mlp_ratio)
    mlp_activation = activation
    self.mlp = Mlp(in_features=dim, hidden_features=mlp_hidden_dim, act_layer=mlp_activation, drop=drop)

    pad = local_conv_size // 2
    self.local_conv = Conv2d_BN(dim, dim, ks=local_conv_size, stride=1, pad=pad, groups=dim)

extra_repr()

TinyViTBlock के मापदंडों का प्रतिनिधित्व करने वाला एक स्वरूपित स्ट्रिंग देता है: आयाम, इनपुट रिज़ॉल्यूशन, की संख्या ध्यान सिर, खिड़की का आकार, और एमएलपी अनुपात।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def extra_repr(self) -> str:
    """Returns a formatted string representing the TinyViTBlock's parameters: dimension, input resolution, number of
    attentions heads, window size, and MLP ratio.
    """
    return (
        f"dim={self.dim}, input_resolution={self.input_resolution}, num_heads={self.num_heads}, "
        f"window_size={self.window_size}, mlp_ratio={self.mlp_ratio}"
    )

forward(x)

स्थानीय के माध्यम से पारित करने से पहले इनपुट 'x' पर ध्यान-आधारित परिवर्तन या पैडिंग लागू करता है कनवल्शन।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def forward(self, x):
    """Applies attention-based transformation or padding to input 'x' before passing it through a local
    convolution.
    """
    h, w = self.input_resolution
    b, l, c = x.shape
    assert l == h * w, "input feature has wrong size"
    res_x = x
    if h == self.window_size and w == self.window_size:
        x = self.attn(x)
    else:
        x = x.view(b, h, w, c)
        pad_b = (self.window_size - h % self.window_size) % self.window_size
        pad_r = (self.window_size - w % self.window_size) % self.window_size
        padding = pad_b > 0 or pad_r > 0

        if padding:
            x = F.pad(x, (0, 0, 0, pad_r, 0, pad_b))

        pH, pW = h + pad_b, w + pad_r
        nH = pH // self.window_size
        nW = pW // self.window_size
        # Window partition
        x = (
            x.view(b, nH, self.window_size, nW, self.window_size, c)
            .transpose(2, 3)
            .reshape(b * nH * nW, self.window_size * self.window_size, c)
        )
        x = self.attn(x)
        # Window reverse
        x = x.view(b, nH, nW, self.window_size, self.window_size, c).transpose(2, 3).reshape(b, pH, pW, c)

        if padding:
            x = x[:, :h, :w].contiguous()

        x = x.view(b, l, c)

    x = res_x + self.drop_path(x)

    x = x.transpose(1, 2).reshape(b, c, h, w)
    x = self.local_conv(x)
    x = x.view(b, c, l).transpose(1, 2)

    return x + self.drop_path(self.mlp(x))



ultralytics.models.sam.modules.tiny_encoder.BasicLayer

का रूप: Module

TinyViT आर्किटेक्चर में एक चरण के लिए एक बुनियादी TinyViT परत।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
class BasicLayer(nn.Module):
    """A basic TinyViT layer for one stage in a TinyViT architecture."""

    def __init__(
        self,
        dim,
        input_resolution,
        depth,
        num_heads,
        window_size,
        mlp_ratio=4.0,
        drop=0.0,
        drop_path=0.0,
        downsample=None,
        use_checkpoint=False,
        local_conv_size=3,
        activation=nn.GELU,
        out_dim=None,
    ):
        """
        Initializes the BasicLayer.

        Args:
            dim (int): The dimensionality of the input and output.
            input_resolution (Tuple[int, int]): Spatial resolution of the input feature map.
            depth (int): Number of TinyViT blocks.
            num_heads (int): Number of attention heads.
            window_size (int): Local window size.
            mlp_ratio (float, optional): Ratio of mlp hidden dim to embedding dim. Default is 4.
            drop (float, optional): Dropout rate. Default is 0.
            drop_path (float | tuple[float], optional): Stochastic depth rate. Default is 0.
            downsample (nn.Module | None, optional): Downsample layer at the end of the layer. Default is None.
            use_checkpoint (bool, optional): Whether to use checkpointing to save memory. Default is False.
            local_conv_size (int, optional): Kernel size of the local convolution. Default is 3.
            activation (torch.nn, optional): Activation function for MLP. Default is nn.GELU.
            out_dim (int | None, optional): The output dimension of the layer. Default is None.

        Raises:
            ValueError: If `drop_path` is a list of float but its length doesn't match `depth`.
        """
        super().__init__()
        self.dim = dim
        self.input_resolution = input_resolution
        self.depth = depth
        self.use_checkpoint = use_checkpoint

        # Build blocks
        self.blocks = nn.ModuleList(
            [
                TinyViTBlock(
                    dim=dim,
                    input_resolution=input_resolution,
                    num_heads=num_heads,
                    window_size=window_size,
                    mlp_ratio=mlp_ratio,
                    drop=drop,
                    drop_path=drop_path[i] if isinstance(drop_path, list) else drop_path,
                    local_conv_size=local_conv_size,
                    activation=activation,
                )
                for i in range(depth)
            ]
        )

        # Patch merging layer
        self.downsample = (
            None
            if downsample is None
            else downsample(input_resolution, dim=dim, out_dim=out_dim, activation=activation)
        )

    def forward(self, x):
        """Performs forward propagation on the input tensor and returns a normalized tensor."""
        for blk in self.blocks:
            x = checkpoint.checkpoint(blk, x) if self.use_checkpoint else blk(x)
        return x if self.downsample is None else self.downsample(x)

    def extra_repr(self) -> str:
        """Returns a string representation of the extra_repr function with the layer's parameters."""
        return f"dim={self.dim}, input_resolution={self.input_resolution}, depth={self.depth}"

__init__(dim, input_resolution, depth, num_heads, window_size, mlp_ratio=4.0, drop=0.0, drop_path=0.0, downsample=None, use_checkpoint=False, local_conv_size=3, activation=nn.GELU, out_dim=None)

BasicLayer को इनिशियलाइज़ करता है।

पैरामीटर:

नाम प्रकार या क़िस्‍म चूक
dim int

इनपुट और आउटपुट की आयामीता।

आवश्यक
input_resolution Tuple[int, int]

इनपुट सुविधा मानचित्र का स्थानिक रिज़ॉल्यूशन।

आवश्यक
depth int

TinyViT ब्लॉकों की संख्या।

आवश्यक
num_heads int

ध्यान प्रमुखों की संख्या।

आवश्यक
window_size int

स्थानीय खिड़की का आकार।

आवश्यक
mlp_ratio float

एमएलपी का अनुपात मंद को एम्बेडिंग मंद करने के लिए छिपा हुआ है। डिफ़ॉल्ट 4 है।

4.0
drop float

ड्रॉपआउट दर। डिफ़ॉल्ट 0 है।

0.0
drop_path float | tuple[float]

स्टोकेस्टिक गहराई दर। डिफ़ॉल्ट 0 है।

0.0
downsample Module | None

परत के अंत में डाउनसैम्पल परत। डिफ़ॉल्ट कोई नहीं है।

None
use_checkpoint bool

स्मृति को बचाने के लिए चेकपॉइंटिंग का उपयोग करना है या नहीं। डिफ़ॉल्ट ग़लत है.

False
local_conv_size int

स्थानीय कनवल्शन का कर्नेल आकार। डिफ़ॉल्ट 3 है।

3
activation nn

एमएलपी के लिए सक्रियण समारोह। डिफ़ॉल्ट nn है. गेलु।

GELU
out_dim int | None

परत का आउटपुट आयाम। डिफ़ॉल्ट कोई नहीं है।

None

उठाती:

प्रकार या क़िस्‍म
ValueError

अगर drop_path फ्लोट की एक सूची है लेकिन इसकी लंबाई मेल नहीं खाती है depth.

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def __init__(
    self,
    dim,
    input_resolution,
    depth,
    num_heads,
    window_size,
    mlp_ratio=4.0,
    drop=0.0,
    drop_path=0.0,
    downsample=None,
    use_checkpoint=False,
    local_conv_size=3,
    activation=nn.GELU,
    out_dim=None,
):
    """
    Initializes the BasicLayer.

    Args:
        dim (int): The dimensionality of the input and output.
        input_resolution (Tuple[int, int]): Spatial resolution of the input feature map.
        depth (int): Number of TinyViT blocks.
        num_heads (int): Number of attention heads.
        window_size (int): Local window size.
        mlp_ratio (float, optional): Ratio of mlp hidden dim to embedding dim. Default is 4.
        drop (float, optional): Dropout rate. Default is 0.
        drop_path (float | tuple[float], optional): Stochastic depth rate. Default is 0.
        downsample (nn.Module | None, optional): Downsample layer at the end of the layer. Default is None.
        use_checkpoint (bool, optional): Whether to use checkpointing to save memory. Default is False.
        local_conv_size (int, optional): Kernel size of the local convolution. Default is 3.
        activation (torch.nn, optional): Activation function for MLP. Default is nn.GELU.
        out_dim (int | None, optional): The output dimension of the layer. Default is None.

    Raises:
        ValueError: If `drop_path` is a list of float but its length doesn't match `depth`.
    """
    super().__init__()
    self.dim = dim
    self.input_resolution = input_resolution
    self.depth = depth
    self.use_checkpoint = use_checkpoint

    # Build blocks
    self.blocks = nn.ModuleList(
        [
            TinyViTBlock(
                dim=dim,
                input_resolution=input_resolution,
                num_heads=num_heads,
                window_size=window_size,
                mlp_ratio=mlp_ratio,
                drop=drop,
                drop_path=drop_path[i] if isinstance(drop_path, list) else drop_path,
                local_conv_size=local_conv_size,
                activation=activation,
            )
            for i in range(depth)
        ]
    )

    # Patch merging layer
    self.downsample = (
        None
        if downsample is None
        else downsample(input_resolution, dim=dim, out_dim=out_dim, activation=activation)
    )

extra_repr()

परत के मापदंडों के साथ extra_repr फ़ंक्शन का एक स्ट्रिंग प्रतिनिधित्व देता है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def extra_repr(self) -> str:
    """Returns a string representation of the extra_repr function with the layer's parameters."""
    return f"dim={self.dim}, input_resolution={self.input_resolution}, depth={self.depth}"

forward(x)

इनपुट पर आगे प्रसार करता है tensor और एक सामान्यीकृत लौटाता है tensor.

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def forward(self, x):
    """Performs forward propagation on the input tensor and returns a normalized tensor."""
    for blk in self.blocks:
        x = checkpoint.checkpoint(blk, x) if self.use_checkpoint else blk(x)
    return x if self.downsample is None else self.downsample(x)



ultralytics.models.sam.modules.tiny_encoder.LayerNorm2d

का रूप: Module

एक PyTorch 2 डी में परत सामान्यीकरण का कार्यान्वयन।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
class LayerNorm2d(nn.Module):
    """A PyTorch implementation of Layer Normalization in 2D."""

    def __init__(self, num_channels: int, eps: float = 1e-6) -> None:
        """Initialize LayerNorm2d with the number of channels and an optional epsilon."""
        super().__init__()
        self.weight = nn.Parameter(torch.ones(num_channels))
        self.bias = nn.Parameter(torch.zeros(num_channels))
        self.eps = eps

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        """Perform a forward pass, normalizing the input tensor."""
        u = x.mean(1, keepdim=True)
        s = (x - u).pow(2).mean(1, keepdim=True)
        x = (x - u) / torch.sqrt(s + self.eps)
        return self.weight[:, None, None] * x + self.bias[:, None, None]

__init__(num_channels, eps=1e-06)

LayerNorm2d को चैनलों की संख्या और एक वैकल्पिक एप्सिलॉन के साथ प्रारंभ करें।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def __init__(self, num_channels: int, eps: float = 1e-6) -> None:
    """Initialize LayerNorm2d with the number of channels and an optional epsilon."""
    super().__init__()
    self.weight = nn.Parameter(torch.ones(num_channels))
    self.bias = nn.Parameter(torch.zeros(num_channels))
    self.eps = eps

forward(x)

इनपुट को सामान्य करते हुए फॉरवर्ड पास करें tensor.

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def forward(self, x: torch.Tensor) -> torch.Tensor:
    """Perform a forward pass, normalizing the input tensor."""
    u = x.mean(1, keepdim=True)
    s = (x - u).pow(2).mean(1, keepdim=True)
    x = (x - u) / torch.sqrt(s + self.eps)
    return self.weight[:, None, None] * x + self.bias[:, None, None]



ultralytics.models.sam.modules.tiny_encoder.TinyViT

का रूप: Module

दृष्टि कार्यों के लिए TinyViT वास्तुकला।

विशेषताएँ:

नाम प्रकार या क़िस्‍म
img_size int

इनपुट छवि का आकार।

in_chans int

इनपुट चैनलों की संख्या।

num_classes int

वर्गीकरण वर्गों की संख्या।

embed_dims List[int]

प्रत्येक परत के लिए एम्बेडिंग आयामों की सूची।

depths List[int]

प्रत्येक परत के लिए गहराई की सूची।

num_heads List[int]

प्रत्येक परत के लिए ध्यान प्रमुखों की संख्या की सूची।

window_sizes List[int]

प्रत्येक परत के लिए खिड़की के आकार की सूची।

mlp_ratio float

एम्बेडिंग आयाम के लिए एमएलपी छिपे हुए आयाम का अनुपात।

drop_rate float

ड्रॉप परतों के लिए ड्रॉपआउट दर।

drop_path_rate float

स्टोकेस्टिक गहराई के लिए ड्रॉप पथ दर।

use_checkpoint bool

कुशल मेमोरी उपयोग के लिए चेकपॉइंटिंग का उपयोग करें।

mbconv_expand_ratio float

MBConv परत के लिए विस्तार अनुपात।

local_conv_size int

स्थानीय कनवल्शन कर्नेल आकार।

layer_lr_decay float

परत-वार सीखने की दर क्षय।

नोट

यह कार्यान्वयन गहराई, ध्यान प्रमुखों की एक सूची को स्वीकार करने के लिए सामान्यीकृत है, आयाम और विंडो आकार एम्बेड करना, जो आपको एक अलग-अलग कॉन्फ़िगरेशन के TinyViT मॉडल का "स्टैक"।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
class TinyViT(nn.Module):
    """
    The TinyViT architecture for vision tasks.

    Attributes:
        img_size (int): Input image size.
        in_chans (int): Number of input channels.
        num_classes (int): Number of classification classes.
        embed_dims (List[int]): List of embedding dimensions for each layer.
        depths (List[int]): List of depths for each layer.
        num_heads (List[int]): List of number of attention heads for each layer.
        window_sizes (List[int]): List of window sizes for each layer.
        mlp_ratio (float): Ratio of MLP hidden dimension to embedding dimension.
        drop_rate (float): Dropout rate for drop layers.
        drop_path_rate (float): Drop path rate for stochastic depth.
        use_checkpoint (bool): Use checkpointing for efficient memory usage.
        mbconv_expand_ratio (float): Expansion ratio for MBConv layer.
        local_conv_size (int): Local convolution kernel size.
        layer_lr_decay (float): Layer-wise learning rate decay.

    Note:
        This implementation is generalized to accept a list of depths, attention heads,
        embedding dimensions and window sizes, which allows you to create a
        "stack" of TinyViT models of varying configurations.
    """

    def __init__(
        self,
        img_size=224,
        in_chans=3,
        num_classes=1000,
        embed_dims=(96, 192, 384, 768),
        depths=(2, 2, 6, 2),
        num_heads=(3, 6, 12, 24),
        window_sizes=(7, 7, 14, 7),
        mlp_ratio=4.0,
        drop_rate=0.0,
        drop_path_rate=0.1,
        use_checkpoint=False,
        mbconv_expand_ratio=4.0,
        local_conv_size=3,
        layer_lr_decay=1.0,
    ):
        """
        Initializes the TinyViT model.

        Args:
            img_size (int, optional): The input image size. Defaults to 224.
            in_chans (int, optional): Number of input channels. Defaults to 3.
            num_classes (int, optional): Number of classification classes. Defaults to 1000.
            embed_dims (List[int], optional): List of embedding dimensions per layer. Defaults to [96, 192, 384, 768].
            depths (List[int], optional): List of depths for each layer. Defaults to [2, 2, 6, 2].
            num_heads (List[int], optional): List of number of attention heads per layer. Defaults to [3, 6, 12, 24].
            window_sizes (List[int], optional): List of window sizes for each layer. Defaults to [7, 7, 14, 7].
            mlp_ratio (float, optional): Ratio of MLP hidden dimension to embedding dimension. Defaults to 4.
            drop_rate (float, optional): Dropout rate. Defaults to 0.
            drop_path_rate (float, optional): Drop path rate for stochastic depth. Defaults to 0.1.
            use_checkpoint (bool, optional): Whether to use checkpointing for efficient memory usage. Defaults to False.
            mbconv_expand_ratio (float, optional): Expansion ratio for MBConv layer. Defaults to 4.0.
            local_conv_size (int, optional): Local convolution kernel size. Defaults to 3.
            layer_lr_decay (float, optional): Layer-wise learning rate decay. Defaults to 1.0.
        """
        super().__init__()
        self.img_size = img_size
        self.num_classes = num_classes
        self.depths = depths
        self.num_layers = len(depths)
        self.mlp_ratio = mlp_ratio

        activation = nn.GELU

        self.patch_embed = PatchEmbed(
            in_chans=in_chans, embed_dim=embed_dims[0], resolution=img_size, activation=activation
        )

        patches_resolution = self.patch_embed.patches_resolution
        self.patches_resolution = patches_resolution

        # Stochastic depth
        dpr = [x.item() for x in torch.linspace(0, drop_path_rate, sum(depths))]  # stochastic depth decay rule

        # Build layers
        self.layers = nn.ModuleList()
        for i_layer in range(self.num_layers):
            kwargs = dict(
                dim=embed_dims[i_layer],
                input_resolution=(
                    patches_resolution[0] // (2 ** (i_layer - 1 if i_layer == 3 else i_layer)),
                    patches_resolution[1] // (2 ** (i_layer - 1 if i_layer == 3 else i_layer)),
                ),
                #   input_resolution=(patches_resolution[0] // (2 ** i_layer),
                #                     patches_resolution[1] // (2 ** i_layer)),
                depth=depths[i_layer],
                drop_path=dpr[sum(depths[:i_layer]) : sum(depths[: i_layer + 1])],
                downsample=PatchMerging if (i_layer < self.num_layers - 1) else None,
                use_checkpoint=use_checkpoint,
                out_dim=embed_dims[min(i_layer + 1, len(embed_dims) - 1)],
                activation=activation,
            )
            if i_layer == 0:
                layer = ConvLayer(conv_expand_ratio=mbconv_expand_ratio, **kwargs)
            else:
                layer = BasicLayer(
                    num_heads=num_heads[i_layer],
                    window_size=window_sizes[i_layer],
                    mlp_ratio=self.mlp_ratio,
                    drop=drop_rate,
                    local_conv_size=local_conv_size,
                    **kwargs,
                )
            self.layers.append(layer)

        # Classifier head
        self.norm_head = nn.LayerNorm(embed_dims[-1])
        self.head = nn.Linear(embed_dims[-1], num_classes) if num_classes > 0 else torch.nn.Identity()

        # Init weights
        self.apply(self._init_weights)
        self.set_layer_lr_decay(layer_lr_decay)
        self.neck = nn.Sequential(
            nn.Conv2d(
                embed_dims[-1],
                256,
                kernel_size=1,
                bias=False,
            ),
            LayerNorm2d(256),
            nn.Conv2d(
                256,
                256,
                kernel_size=3,
                padding=1,
                bias=False,
            ),
            LayerNorm2d(256),
        )

    def set_layer_lr_decay(self, layer_lr_decay):
        """Sets the learning rate decay for each layer in the TinyViT model."""
        decay_rate = layer_lr_decay

        # Layers -> blocks (depth)
        depth = sum(self.depths)
        lr_scales = [decay_rate ** (depth - i - 1) for i in range(depth)]

        def _set_lr_scale(m, scale):
            """Sets the learning rate scale for each layer in the model based on the layer's depth."""
            for p in m.parameters():
                p.lr_scale = scale

        self.patch_embed.apply(lambda x: _set_lr_scale(x, lr_scales[0]))
        i = 0
        for layer in self.layers:
            for block in layer.blocks:
                block.apply(lambda x: _set_lr_scale(x, lr_scales[i]))
                i += 1
            if layer.downsample is not None:
                layer.downsample.apply(lambda x: _set_lr_scale(x, lr_scales[i - 1]))
        assert i == depth
        for m in [self.norm_head, self.head]:
            m.apply(lambda x: _set_lr_scale(x, lr_scales[-1]))

        for k, p in self.named_parameters():
            p.param_name = k

        def _check_lr_scale(m):
            """Checks if the learning rate scale attribute is present in module's parameters."""
            for p in m.parameters():
                assert hasattr(p, "lr_scale"), p.param_name

        self.apply(_check_lr_scale)

    def _init_weights(self, m):
        """Initializes weights for linear layers and layer normalization in the given module."""
        if isinstance(m, nn.Linear):
            # NOTE: This initialization is needed only for training.
            # trunc_normal_(m.weight, std=.02)
            if m.bias is not None:
                nn.init.constant_(m.bias, 0)
        elif isinstance(m, nn.LayerNorm):
            nn.init.constant_(m.bias, 0)
            nn.init.constant_(m.weight, 1.0)

    @torch.jit.ignore
    def no_weight_decay_keywords(self):
        """Returns a dictionary of parameter names where weight decay should not be applied."""
        return {"attention_biases"}

    def forward_features(self, x):
        """Runs the input through the model layers and returns the transformed output."""
        x = self.patch_embed(x)  # x input is (N, C, H, W)

        x = self.layers[0](x)
        start_i = 1

        for i in range(start_i, len(self.layers)):
            layer = self.layers[i]
            x = layer(x)
        batch, _, channel = x.shape
        x = x.view(batch, 64, 64, channel)
        x = x.permute(0, 3, 1, 2)
        return self.neck(x)

    def forward(self, x):
        """Executes a forward pass on the input tensor through the constructed model layers."""
        return self.forward_features(x)

__init__(img_size=224, in_chans=3, num_classes=1000, embed_dims=(96, 192, 384, 768), depths=(2, 2, 6, 2), num_heads=(3, 6, 12, 24), window_sizes=(7, 7, 14, 7), mlp_ratio=4.0, drop_rate=0.0, drop_path_rate=0.1, use_checkpoint=False, mbconv_expand_ratio=4.0, local_conv_size=3, layer_lr_decay=1.0)

TinyViT मॉडल को इनिशियलाइज़ करता है।

पैरामीटर:

नाम प्रकार या क़िस्‍म चूक
img_size int

इनपुट छवि का आकार। 224 के लिए डिफ़ॉल्ट।

224
in_chans int

इनपुट चैनलों की संख्या। 3 के लिए डिफ़ॉल्ट।

3
num_classes int

वर्गीकरण वर्गों की संख्या। 1000 के लिए डिफ़ॉल्ट।

1000
embed_dims List[int]

प्रति परत एम्बेडिंग आयामों की सूची। [96, 192, 384, 768] के लिए डिफ़ॉल्ट।

(96, 192, 384, 768)
depths List[int]

प्रत्येक परत के लिए गहराई की सूची। [2, 2, 6, 2] के लिए डिफ़ॉल्ट।

(2, 2, 6, 2)
num_heads List[int]

प्रति परत ध्यान सिर की संख्या की सूची। [3, 6, 12, 24] के लिए डिफ़ॉल्ट।

(3, 6, 12, 24)
window_sizes List[int]

प्रत्येक परत के लिए खिड़की के आकार की सूची। [7, 7, 14, 7] के लिए डिफ़ॉल्ट।

(7, 7, 14, 7)
mlp_ratio float

एम्बेडिंग आयाम के लिए एमएलपी छिपे हुए आयाम का अनुपात। 4 के लिए डिफ़ॉल्ट।

4.0
drop_rate float

ड्रॉपआउट दर। 0 के लिए डिफ़ॉल्ट।

0.0
drop_path_rate float

स्टोकेस्टिक गहराई के लिए ड्रॉप पथ दर। 0.1 के लिए डिफ़ॉल्ट।

0.1
use_checkpoint bool

कुशल मेमोरी उपयोग के लिए चेकपॉइंटिंग का उपयोग करना है या नहीं। डिफ़ॉल्ट रूप से गलत है.

False
mbconv_expand_ratio float

MBConv परत के लिए विस्तार अनुपात। 4.0 के लिए डिफ़ॉल्ट।

4.0
local_conv_size int

स्थानीय कनवल्शन कर्नेल आकार। 3 के लिए डिफ़ॉल्ट।

3
layer_lr_decay float

परत-वार सीखने की दर क्षय। 1.0 के लिए डिफ़ॉल्ट।

1.0
में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def __init__(
    self,
    img_size=224,
    in_chans=3,
    num_classes=1000,
    embed_dims=(96, 192, 384, 768),
    depths=(2, 2, 6, 2),
    num_heads=(3, 6, 12, 24),
    window_sizes=(7, 7, 14, 7),
    mlp_ratio=4.0,
    drop_rate=0.0,
    drop_path_rate=0.1,
    use_checkpoint=False,
    mbconv_expand_ratio=4.0,
    local_conv_size=3,
    layer_lr_decay=1.0,
):
    """
    Initializes the TinyViT model.

    Args:
        img_size (int, optional): The input image size. Defaults to 224.
        in_chans (int, optional): Number of input channels. Defaults to 3.
        num_classes (int, optional): Number of classification classes. Defaults to 1000.
        embed_dims (List[int], optional): List of embedding dimensions per layer. Defaults to [96, 192, 384, 768].
        depths (List[int], optional): List of depths for each layer. Defaults to [2, 2, 6, 2].
        num_heads (List[int], optional): List of number of attention heads per layer. Defaults to [3, 6, 12, 24].
        window_sizes (List[int], optional): List of window sizes for each layer. Defaults to [7, 7, 14, 7].
        mlp_ratio (float, optional): Ratio of MLP hidden dimension to embedding dimension. Defaults to 4.
        drop_rate (float, optional): Dropout rate. Defaults to 0.
        drop_path_rate (float, optional): Drop path rate for stochastic depth. Defaults to 0.1.
        use_checkpoint (bool, optional): Whether to use checkpointing for efficient memory usage. Defaults to False.
        mbconv_expand_ratio (float, optional): Expansion ratio for MBConv layer. Defaults to 4.0.
        local_conv_size (int, optional): Local convolution kernel size. Defaults to 3.
        layer_lr_decay (float, optional): Layer-wise learning rate decay. Defaults to 1.0.
    """
    super().__init__()
    self.img_size = img_size
    self.num_classes = num_classes
    self.depths = depths
    self.num_layers = len(depths)
    self.mlp_ratio = mlp_ratio

    activation = nn.GELU

    self.patch_embed = PatchEmbed(
        in_chans=in_chans, embed_dim=embed_dims[0], resolution=img_size, activation=activation
    )

    patches_resolution = self.patch_embed.patches_resolution
    self.patches_resolution = patches_resolution

    # Stochastic depth
    dpr = [x.item() for x in torch.linspace(0, drop_path_rate, sum(depths))]  # stochastic depth decay rule

    # Build layers
    self.layers = nn.ModuleList()
    for i_layer in range(self.num_layers):
        kwargs = dict(
            dim=embed_dims[i_layer],
            input_resolution=(
                patches_resolution[0] // (2 ** (i_layer - 1 if i_layer == 3 else i_layer)),
                patches_resolution[1] // (2 ** (i_layer - 1 if i_layer == 3 else i_layer)),
            ),
            #   input_resolution=(patches_resolution[0] // (2 ** i_layer),
            #                     patches_resolution[1] // (2 ** i_layer)),
            depth=depths[i_layer],
            drop_path=dpr[sum(depths[:i_layer]) : sum(depths[: i_layer + 1])],
            downsample=PatchMerging if (i_layer < self.num_layers - 1) else None,
            use_checkpoint=use_checkpoint,
            out_dim=embed_dims[min(i_layer + 1, len(embed_dims) - 1)],
            activation=activation,
        )
        if i_layer == 0:
            layer = ConvLayer(conv_expand_ratio=mbconv_expand_ratio, **kwargs)
        else:
            layer = BasicLayer(
                num_heads=num_heads[i_layer],
                window_size=window_sizes[i_layer],
                mlp_ratio=self.mlp_ratio,
                drop=drop_rate,
                local_conv_size=local_conv_size,
                **kwargs,
            )
        self.layers.append(layer)

    # Classifier head
    self.norm_head = nn.LayerNorm(embed_dims[-1])
    self.head = nn.Linear(embed_dims[-1], num_classes) if num_classes > 0 else torch.nn.Identity()

    # Init weights
    self.apply(self._init_weights)
    self.set_layer_lr_decay(layer_lr_decay)
    self.neck = nn.Sequential(
        nn.Conv2d(
            embed_dims[-1],
            256,
            kernel_size=1,
            bias=False,
        ),
        LayerNorm2d(256),
        nn.Conv2d(
            256,
            256,
            kernel_size=3,
            padding=1,
            bias=False,
        ),
        LayerNorm2d(256),
    )

forward(x)

इनपुट पर एक फॉरवर्ड पास निष्पादित करता है tensor निर्मित मॉडल परतों के माध्यम से।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def forward(self, x):
    """Executes a forward pass on the input tensor through the constructed model layers."""
    return self.forward_features(x)

forward_features(x)

मॉडल परतों के माध्यम से इनपुट चलाता है और रूपांतरित आउटपुट लौटाता है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def forward_features(self, x):
    """Runs the input through the model layers and returns the transformed output."""
    x = self.patch_embed(x)  # x input is (N, C, H, W)

    x = self.layers[0](x)
    start_i = 1

    for i in range(start_i, len(self.layers)):
        layer = self.layers[i]
        x = layer(x)
    batch, _, channel = x.shape
    x = x.view(batch, 64, 64, channel)
    x = x.permute(0, 3, 1, 2)
    return self.neck(x)

no_weight_decay_keywords()

पैरामीटर नामों का एक शब्दकोश देता है जहां वजन क्षय लागू नहीं किया जाना चाहिए।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
@torch.jit.ignore
def no_weight_decay_keywords(self):
    """Returns a dictionary of parameter names where weight decay should not be applied."""
    return {"attention_biases"}

set_layer_lr_decay(layer_lr_decay)

TinyViT मॉडल में प्रत्येक परत के लिए सीखने की दर क्षय सेट करता है।

में स्रोत कोड ultralytics/models/sam/modules/tiny_encoder.py
def set_layer_lr_decay(self, layer_lr_decay):
    """Sets the learning rate decay for each layer in the TinyViT model."""
    decay_rate = layer_lr_decay

    # Layers -> blocks (depth)
    depth = sum(self.depths)
    lr_scales = [decay_rate ** (depth - i - 1) for i in range(depth)]

    def _set_lr_scale(m, scale):
        """Sets the learning rate scale for each layer in the model based on the layer's depth."""
        for p in m.parameters():
            p.lr_scale = scale

    self.patch_embed.apply(lambda x: _set_lr_scale(x, lr_scales[0]))
    i = 0
    for layer in self.layers:
        for block in layer.blocks:
            block.apply(lambda x: _set_lr_scale(x, lr_scales[i]))
            i += 1
        if layer.downsample is not None:
            layer.downsample.apply(lambda x: _set_lr_scale(x, lr_scales[i - 1]))
    assert i == depth
    for m in [self.norm_head, self.head]:
        m.apply(lambda x: _set_lr_scale(x, lr_scales[-1]))

    for k, p in self.named_parameters():
        p.param_name = k

    def _check_lr_scale(m):
        """Checks if the learning rate scale attribute is present in module's parameters."""
        for p in m.parameters():
            assert hasattr(p, "lr_scale"), p.param_name

    self.apply(_check_lr_scale)





Created 2023-11-12, Updated 2024-06-02
Authors: glenn-jocher (5), Burhan-Q (1), Laughing-q (1)