笔者碎碎念：Swin-T作为ICCV21的最佳论文，一直没来得及看代码，今天就着假期看看代码学习。本文不涉及任何公式推导，并且要求读者有ViT的基础知识，如知道什么是MHSA等。 并且限于笔者水平，可能会有些错误，还请各位看官怒斥。 2022/4/4

codebase: https://github.com/microsoft/Swin-Transformer/blob/main/models/swin_transformer.py

attention公式镇场子

总体架构
Swin-T的结构有下面几个：Patch Patition, Patch Merging, Swin Transformer Block
其中最重要的Swin Transformer Block将ViT的MHSA变成了W-MSA和SW-MSA（一种带shift的，一种不带是shift的）。

总体架构

class SwinTransformer(nn.Module):
    r""" Swin Transformer
        A PyTorch impl of : `Swin Transformer: Hierarchical Vision Transformer using Shifted Windows`  -
          https://arxiv.org/pdf/2103.14030
    Args:
        img_size (int | tuple(int)): Input image size. Default 224
        patch_size (int | tuple(int)): Patch size. Default: 4
        in_chans (int): Number of input image channels. Default: 3
        num_classes (int): Number of classes for classification head. Default: 1000
        embed_dim (int): Patch embedding dimension. Default: 96
        depths (tuple(int)): Depth of each Swin Transformer layer.
        num_heads (tuple(int)): Number of attention heads in different layers.
        window_size (int): Window size. Default: 7
        mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. Default: 4
        qkv_bias (bool): If True, add a learnable bias to query, key, value. Default: True
        qk_scale (float): Override default qk scale of head_dim ** -0.5 if set. Default: None
        drop_rate (float): Dropout rate. Default: 0
        attn_drop_rate (float): Attention dropout rate. Default: 0
        drop_path_rate (float): Stochastic depth rate. Default: 0.1
        norm_layer (nn.Module): Normalization layer. Default: nn.LayerNorm.
        ape (bool): If True, add absolute position embedding to the patch embedding. Default: False
        patch_norm (bool): If True, add normalization after patch embedding. Default: True
        use_checkpoint (bool): Whether to use checkpointing to save memory. Default: False
    """
    def __init__(self, img_size=224, patch_size=4, in_chans=3, num_classes=1000,
                 embed_dim=96, depths=[2, 2, 6, 2], num_heads=[3, 6, 12, 24],
                 window_size=7, mlp_ratio=4., qkv_bias=True, qk_scale=None,
                 drop_rate=0., attn_drop_rate=0., drop_path_rate=0.1,
                 norm_layer=nn.LayerNorm, ape=False, patch_norm=True,
                 use_checkpoint=False, **kwargs):
        super().__init__()
        self.num_classes = num_classes
        self.num_layers = len(depths)
        self.embed_dim = embed_dim
        self.ape = ape
        self.patch_norm = patch_norm
        self.num_features = int(embed_dim * 2 ** (self.num_layers - 1))
        self.mlp_ratio = mlp_ratio
        # split image into non-overlapping patches
        self.patch_embed = PatchEmbed(
            img_size=img_size, patch_size=patch_size, in_chans=in_chans, embed_dim=embed_dim,
            norm_layer=norm_layer if self.patch_norm else None)
        num_patches = self.patch_embed.num_patches
        patches_resolution = self.patch_embed.patches_resolution
        self.patches_resolution = patches_resolution
        # absolute position embedding
        if self.ape:
            self.absolute_pos_embed = nn.Parameter(torch.zeros(1, num_patches, embed_dim))
            trunc_normal_(self.absolute_pos_embed, std=.02)
        self.pos_drop = nn.Dropout(p=drop_rate)
        # 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):
            layer = BasicLayer(dim=int(embed_dim * 2 ** i_layer),
                               input_resolution=(patches_resolution[0] // (2 ** i_layer),
                                                 patches_resolution[1] // (2 ** i_layer)),
                               depth=depths[i_layer],
                               num_heads=num_heads[i_layer],
                               window_size=window_size,
                               mlp_ratio=self.mlp_ratio,
                               qkv_bias=qkv_bias, qk_scale=qk_scale,
                               drop=drop_rate, attn_drop=attn_drop_rate,
                               drop_path=dpr[sum(depths[:i_layer]):sum(depths[:i_layer + 1])],
                               norm_layer=norm_layer,
                               downsample=PatchMerging if (i_layer < self.num_layers - 1) else None,
                               use_checkpoint=use_checkpoint)
            self.layers.append(layer)
        self.norm = norm_layer(self.num_features)
        self.avgpool = nn.AdaptiveAvgPool1d(1)
        self.head = nn.Linear(self.num_features, num_classes) if num_classes > 0 else nn.Identity()
        self.apply(self._init_weights)
    def _init_weights(self, m):
        if isinstance(m, nn.Linear):
            trunc_normal_(m.weight, std=.02)
            if isinstance(m, nn.Linear) and 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)
    def forward_features(self, x):
        x = self.patch_embed(x) # patchify
        if self.ape:
            x = x + self.absolute_pos_embed
        x = self.pos_drop(x)
        for layer in self.layers:
            x = layer(x)
        x = self.norm(x)  # B L C
        x = self.avgpool(x.transpose(1, 2))  # B C 1
        x = torch.flatten(x, 1)
        return x
    def forward(self, x):
        x = self.forward_features(x)
        x = self.head(x)
        return

从forward中看出，图像进去PatchEmbed层先将其划分为一个个token，再加上positional embedding（假设这里默认是绝对位置嵌入），随后是dropout，然后是每个layer层计算注意力，最后将输出展平输出1000类logits（假设是ImageNet图片分类）。

所以我们需要看的就是BasicLayer层如何定义。

其他层

在看BasicLayer如何定义之前，先给出一些简单的helper layers

Mlp

class Mlp(nn.Module):
    def __init__(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.):
        super().__init__()
        out_features = out_features or in_features
        hidden_features = hidden_features or in_features
        self.fc1 = nn.Linear(in_features, hidden_features)
        self.act = act_layer()
        self.fc2 = nn.Linear(hidden_features, out_features)
        self.drop = nn.Dropout(drop)
    def forward(self, x):
        x = self.fc1(x)
        x = self.act(x)
        x = self.drop(x)
        x = self.fc2(x)
        x = self.drop(x)
        return x

2层mlp不多说，大家都知道。

Window Partition&Reverse

def window_partition(x, window_size):
    """
    Args:
        x: (B, H, W, C)
        window_size (int): window size
    Returns:
        windows: (num_windows*B, window_size, window_size, C)
    """
    B, H, W, C = x.shape
    x = x.view(B, H // window_size, window_size, W // window_size, window_size, C)
    windows = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(-1, window_size, window_size, C)
    return windows
def window_reverse(windows, window_size, H, W):
    """
    Args:
        windows: (num_windows*B, window_size, window_size, C)
        window_size (int): Window size
        H (int): Height of image
        W (int): Width of image
    Returns:
        x: (B, H, W, C)
    """
    B = int(windows.shape[0] / (H * W / window_size / window_size))
    x = windows.view(B, H // window_size, W // window_size, window_size, window_size, -1)
    x = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(B, H, W, -1)
    return x

这里的维数有点多，需要大家仔细看。图片进来，直接将其划分为一个个window_size大小的patch。或者，token进来变成图片。

PatchEmbed&PatchMerge

PatchEmbed

class PatchEmbed(nn.Module):
    r""" Image to Patch Embedding
    Args:
        img_size (int): Image size.  Default: 224.
        patch_size (int): Patch token size. Default: 4.
        in_chans (int): Number of input image channels. Default: 3.
        embed_dim (int): Number of linear projection output channels. Default: 96.
        norm_layer (nn.Module, optional): Normalization layer. Default: None
    """
    def __init__(self, img_size=224, patch_size=4, in_chans=3, embed_dim=96, norm_layer=None):
        super().__init__()
        img_size = to_2tuple(img_size)
        patch_size = to_2tuple(patch_size)
        patches_resolution = [img_size[0] // patch_size[0], img_size[1] // patch_size[1]]
        self.img_size = img_size
        self.patch_size = patch_size
        self.patches_resolution = patches_resolution
        self.num_patches = patches_resolution[0] * patches_resolution[1]
        self.in_chans = in_chans
        self.embed_dim = embed_dim
        self.proj = nn.Conv2d(in_chans, embed_dim, kernel_size=patch_size, stride=patch_size)
        if norm_layer is not None:
            self.norm = norm_layer(embed_dim)
        else:
            self.norm = None
    def forward(self, x):
        B, C, H, W = x.shape
        # FIXME look at relaxing size constraints
        assert H == self.img_size[0] and W == self.img_size[1], \
            f"Input image size ({H}*{W}) doesn't match model ({self.img_size[0]}*{self.img_size[1]})."
        x = self.proj(x).flatten(2).transpose(1, 2)  # B Ph*Pw C
        if self.norm is not None:
            x = self.norm(x)
        return x

PatchEmbed层对于进来的图片划分为一个个的patch，具体使用kernel_size和stride相等的卷积操作，输入维度是96，这样就完成了patchify。对于ImageNet的图片，这一层的输出为(B, 56*56, 96)，这样就可以和positional_embedding相加了（x = x + self.absolute_pos_embed)

PatchMerging

class PatchMerging(nn.Module):
    r""" Patch Merging Layer.
    Args:
        input_resolution (tuple[int]): Resolution of input feature.
        dim (int): Number of input channels.
        norm_layer (nn.Module, optional): Normalization layer.  Default: nn.LayerNorm
    """
    def __init__(self, input_resolution, dim, norm_layer=nn.LayerNorm):
        super().__init__()
        self.input_resolution = input_resolution
        self.dim = dim
        self.reduction = nn.Linear(4 * dim, 2 * dim, bias=False)
        self.norm = norm_layer(4 * dim)
    def forward(self, x):
        """
        x: B, H*W, C
        """
        H, W = self.input_resolution
        B, L, C = x.shape
        assert L == H * W, "input feature has wrong size"
        assert H % 2 == 0 and W % 2 == 0, f"x size ({H}*{W}) are not even."
        x = x.view(B, H, W, C)
        x0 = x[:, 0::2, 0::2, :]  # B H/2 W/2 C
        x1 = x[:, 1::2, 0::2, :]  # B H/2 W/2 C
        x2 = x[:, 0::2, 1::2, :]  # B H/2 W/2 C
        x3 = x[:, 1::2, 1::2, :]  # B H/2 W/2 C
        x = torch.cat([x0, x1, x2, x3], -1)  # B H/2 W/2 4*C
        x = x.view(B, -1, 4 * C)  # B H/2*W/2 4*C
        x = self.norm(x)
        x = self.reduction(x)
        return x

PatchMerging层的作用是将图片的长宽减半，通道数增加2倍。

图源自知乎

很像Pixshuffle的反操作

BasicLayer

好啦，一些基础的层到这里就结束了。可以开始重点了。

class BasicLayer(nn.Module):
    """ A basic Swin Transformer layer for one stage.
    Args:
        dim (int): Number of input channels.
        input_resolution (tuple[int]): Input resolution.
        depth (int): Number of blocks.
        num_heads (int): Number of attention heads.
        window_size (int): Local window size.
        mlp_ratio (float): Ratio of mlp hidden dim to embedding dim.
        qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True
        qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set.
        drop (float, optional): Dropout rate. Default: 0.0
        attn_drop (float, optional): Attention dropout rate. Default: 0.0
        drop_path (float | tuple[float], optional): Stochastic depth rate. Default: 0.0
        norm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNorm
        downsample (nn.Module | None, optional): Downsample layer at the end of the layer. Default: None
        use_checkpoint (bool): Whether to use checkpointing to save memory. Default: False.
    """
    def __init__(self, dim, input_resolution, depth, num_heads, window_size,
                 mlp_ratio=4., qkv_bias=True, qk_scale=None, drop=0., attn_drop=0.,
                 drop_path=0., norm_layer=nn.LayerNorm, downsample=None, use_checkpoint=False):
        super().__init__()
        self.dim = dim
        self.input_resolution = input_resolution
        self.depth = depth
        self.use_checkpoint = use_checkpoint
        # build blocks
        self.blocks = nn.ModuleList([
            SwinTransformerBlock(dim=dim, input_resolution=input_resolution,
                                 num_heads=num_heads, window_size=window_size,
                                 shift_size=0 if (i % 2 == 0) else window_size // 2,
                                 mlp_ratio=mlp_ratio,
                                 qkv_bias=qkv_bias, qk_scale=qk_scale,
                                 drop=drop, attn_drop=attn_drop,
                                 drop_path=drop_path[i] if isinstance(drop_path, list) else drop_path,
                                 norm_layer=norm_layer)
            for i in range(depth)])
        # patch merging layer
        if downsample is not None:
            self.downsample = downsample(input_resolution, dim=dim, norm_layer=norm_layer)
        else:
            self.downsample = None
    def forward(self, x):
        for blk in self.blocks:
            if self.use_checkpoint:
                x = checkpoint.checkpoint(blk, x)
            else:
                x = blk(x)
        if self.downsample is not None:
            x = self.downsample(x)
        return

BasicLayer层中包含了很多层SwinTransformerBlock，根据你给定的深度，创建depth个Block。需要的话，每个Block后还会有PatchMerging层。（套娃是吧）

SwinTransformerBlock

既然套娃那就接着看。

class SwinTransformerBlock(nn.Module):
    r""" Swin Transformer Block.
    Args:
        dim (int): Number of input channels.
        input_resolution (tuple[int]): Input resulotion.
        num_heads (int): Number of attention heads.
        window_size (int): Window size.
        shift_size (int): Shift size for SW-MSA.
        mlp_ratio (float): Ratio of mlp hidden dim to embedding dim.
        qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True
        qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set.
        drop (float, optional): Dropout rate. Default: 0.0
        attn_drop (float, optional): Attention dropout rate. Default: 0.0
        drop_path (float, optional): Stochastic depth rate. Default: 0.0
        act_layer (nn.Module, optional): Activation layer. Default: nn.GELU
        norm_layer (nn.Module, optional): Normalization layer.  Default: nn.LayerNorm
    """

    def __init__(self, dim, input_resolution, num_heads, window_size=7, shift_size=0,
                 mlp_ratio=4., qkv_bias=True, qk_scale=None, drop=0., attn_drop=0., drop_path=0.,
                 act_layer=nn.GELU, norm_layer=nn.LayerNorm):
        super().__init__()
        self.dim = dim
        self.input_resolution = input_resolution
        self.num_heads = num_heads
        self.window_size = window_size
        self.shift_size = shift_size
        self.mlp_ratio = mlp_ratio
        if min(self.input_resolution) <= self.window_size:
            # if window size is larger than input resolution, we don't partition windows
            self.shift_size = 0
            self.window_size = min(self.input_resolution)
        assert 0 <= self.shift_size < self.window_size, "shift_size must in 0-window_size"

        self.norm1 = norm_layer(dim)
        self.attn = WindowAttention(
            dim, window_size=to_2tuple(self.window_size), num_heads=num_heads,
            qkv_bias=qkv_bias, qk_scale=qk_scale, attn_drop=attn_drop, proj_drop=drop)

        self.drop_path = DropPath(drop_path) if drop_path > 0. else nn.Identity()
        self.norm2 = norm_layer(dim)
        mlp_hidden_dim = int(dim * mlp_ratio)
        self.mlp = Mlp(in_features=dim, hidden_features=mlp_hidden_dim, act_layer=act_layer, drop=drop)

        if self.shift_size > 0:
            # calculate attention mask for SW-MSA
            H, W = self.input_resolution
            img_mask = torch.zeros((1, H, W, 1))  # 1 H W 1
            h_slices = (slice(0, -self.window_size),
                        slice(-self.window_size, -self.shift_size),
                        slice(-self.shift_size, None))
            w_slices = (slice(0, -self.window_size),
                        slice(-self.window_size, -self.shift_size),
                        slice(-self.shift_size, None))
            cnt = 0
            for h in h_slices:
                for w in w_slices:
                    img_mask[:, h, w, :] = cnt
                    cnt += 1

            mask_windows = window_partition(img_mask, self.window_size)  # nW, window_size, window_size, 1
            mask_windows = mask_windows.view(-1, self.window_size * self.window_size)
            attn_mask = mask_windows.unsqueeze(1) - mask_windows.unsqueeze(2)
            attn_mask = attn_mask.masked_fill(attn_mask != 0, float(-100.0)).masked_fill(attn_mask == 0, float(0.0))
        else:
            attn_mask = None

        self.register_buffer("attn_mask", attn_mask)

    def forward(self, x):
        H, W = self.input_resolution
        B, L, C = x.shape
        assert L == H * W, "input feature has wrong size"

        shortcut = x
        x = self.norm1(x)
        x = x.view(B, H, W, C)

        # cyclic shift
        if self.shift_size > 0:
            shifted_x = torch.roll(x, shifts=(-self.shift_size, -self.shift_size), dims=(1, 2))
        else:
            shifted_x = x

        # partition windows
        x_windows = window_partition(shifted_x, self.window_size)  # nW*B, window_size, window_size, C
        x_windows = x_windows.view(-1, self.window_size * self.window_size, C)  # nW*B, window_size*window_size, C

        # W-MSA/SW-MSA
        attn_windows = self.attn(x_windows, mask=self.attn_mask)  # nW*B, window_size*window_size, C

        # merge windows
        attn_windows = attn_windows.view(-1, self.window_size, self.window_size, C)
        shifted_x = window_reverse(attn_windows, self.window_size, H, W)  # B H' W' C

        # reverse cyclic shift
        if self.shift_size > 0:
            x = torch.roll(shifted_x, shifts=(self.shift_size, self.shift_size), dims=(1, 2))
        else:
            x = shifted_x
        x = x.view(B, H * W, C)

        # FFN
        x = shortcut + self.drop_path(x)
        x = x + self.drop_path(self.mlp(self.norm2(x)))

        return x

这一层就是W-MSA和SW-MSA层啦，通过shift_size决定是否shift以及shift多少。

先看forward，我们先忽略是否shift的问题，SwinTransformerBlock首先将输入的x划分为很多个window，attention操作就是在这些window里进行，这样可以减少计算量（可以理解为local attention？），输出的维度为(num_window*B, window_size*window_size, C)

接着就做attention操作了，代码里又另起一层专门写这个attention（恼）。

然后将每个window里做好的attention重新reverse成图片的形式(B, H', W', C)

随后过norm和Mlp还有shortcut，这样，完整的Block的流程就走完了。

WindowAttention

接着就是喜闻乐见的Attention环节，熟悉ViT的朋友们看这一节肯定会非常容易。

class WindowAttention(nn.Module):
    r""" Window based multi-head self attention (W-MSA) module with relative position bias.
    It supports both of shifted and non-shifted window.
    Args:
        dim (int): Number of input channels.
        window_size (tuple[int]): The height and width of the window.
        num_heads (int): Number of attention heads.
        qkv_bias (bool, optional):  If True, add a learnable bias to query, key, value. Default: True
        qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set
        attn_drop (float, optional): Dropout ratio of attention weight. Default: 0.0
        proj_drop (float, optional): Dropout ratio of output. Default: 0.0
    """

    def __init__(self, dim, window_size, num_heads, qkv_bias=True, qk_scale=None, attn_drop=0., proj_drop=0.):

        super().__init__()
        self.dim = dim
        self.window_size = window_size  # Wh, Ww
        self.num_heads = num_heads
        head_dim = dim // num_heads
        self.scale = qk_scale or head_dim ** -0.5

        # define a parameter table of relative position bias
        self.relative_position_bias_table = nn.Parameter(
            torch.zeros((2 * window_size[0] - 1) * (2 * window_size[1] - 1), num_heads))  # 2*Wh-1 * 2*Ww-1, nH
            # 2*wh-1 is max(relative_position_index)?

        # get pair-wise relative position index for each token inside the window
        coords_h = torch.arange(self.window_size[0])
        coords_w = torch.arange(self.window_size[1])
        coords = torch.stack(torch.meshgrid([coords_h, coords_w]))  # 2, Wh, Ww
        coords_flatten = torch.flatten(coords, 1)  # 2, Wh*Ww
        relative_coords = coords_flatten[:, :, None] - coords_flatten[:, None, :]  # 2, Wh*Ww, Wh*Ww
        relative_coords = relative_coords.permute(1, 2, 0).contiguous()  # Wh*Ww, Wh*Ww, 2
        relative_coords[:, :, 0] += self.window_size[0] - 1  # shift to start from 0
        relative_coords[:, :, 1] += self.window_size[1] - 1
        relative_coords[:, :, 0] *= 2 * self.window_size[1] - 1
        relative_position_index = relative_coords.sum(-1)  # Wh*Ww, Wh*Ww
        self.register_buffer("relative_position_index", relative_position_index)

        self.qkv = nn.Linear(dim, dim * 3, bias=qkv_bias)
        self.attn_drop = nn.Dropout(attn_drop)
        self.proj = nn.Linear(dim, dim)
        self.proj_drop = nn.Dropout(proj_drop)

        trunc_normal_(self.relative_position_bias_table, std=.02)
        self.softmax = nn.Softmax(dim=-1)

    def forward(self, x, mask=None):
        """
        Args:
            x: input features with shape of (num_windows*B, N, C)
            mask: (0/-inf) mask with shape of (num_windows, Wh*Ww, Wh*Ww) or None
        """
        B_, N, C = x.shape
        qkv = self.qkv(x).reshape(B_, N, 3, self.num_heads, C // self.num_heads).permute(2, 0, 3, 1, 4)
        q, k, v = qkv[0], qkv[1], qkv[2]  # make torchscript happy (cannot use tensor as tuple)
        # q size: [B*nw, wh*ww, C//h] N=wh*ww

        q = q * self.scale
        attn = (q @ k.transpose(-2, -1)) # [B*nw, N, N]

        relative_position_bias = self.relative_position_bias_table[self.relative_position_index.view(-1)].view(
            self.window_size[0] * self.window_size[1], self.window_size[0] * self.window_size[1], -1)  # Wh*Ww,Wh*Ww,nH
        relative_position_bias = relative_position_bias.permute(2, 0, 1).contiguous()  # nH, Wh*Ww, Wh*Ww
        attn = attn + relative_position_bias.unsqueeze(0)

        if mask is not None:
            nW = mask.shape[0]
            attn = attn.view(B_ // nW, nW, self.num_heads, N, N) + mask.unsqueeze(1).unsqueeze(0)
            attn = attn.view(-1, self.num_heads, N, N)
            attn = self.softmax(attn)
        else:
            attn = self.softmax(attn)

        attn = self.attn_drop(attn)

        x = (attn @ v).transpose(1, 2).reshape(B_, N, C)
        x = self.proj(x)
        x = self.proj_drop(x)
        return x

对于forward的前几行，就是常规的attention操作（计算出q, k, v，然后按照公式去做就完事了）。

然后后面就多了个relative_position_bias与计算出的attention相加，这个东西叫相对位置编码，实验证明有助于提升模型性能。
这里的代码有点复杂，我们一点点来捋。

我们这里以(2,2)的window_size为例，首先是coords：

coords_h = torch.arange(self.window_size[0])
coords_w = torch.arange(self.window_size[1])
coords = torch.stack(torch.meshgrid([coords_h, coords_w]))  # 2, Wh, Ww
coords_flatten = torch.flatten(coords, 1)  # 2, Wh*Ww

输出为

tensor([[[0, 0],
          [1, 1]],

         [[0, 1],
          [0, 1]]]))

flatten展平得到coords_flatten

tensor([[0, 0, 1, 1],
         [0, 1, 0, 1]])

利用广播机制相减得到relative_coords，再将其平移至0，再对其中的某一维做乘法区分

relative_coords = coords_flatten[:, :, None] - coords_flatten[:, None, :]  # 2, Wh*Ww, Wh*Ww
relative_coords = relative_coords.permute(1, 2, 0).contiguous()  # Wh*Ww, Wh*Ww, 2
relative_coords[:, :, 0] += self.window_size[0] - 1  # shift to start from 0
relative_coords[:, :, 1] += self.window_size[1] - 1

最后再对最后一维相加得到relative_position_index

relative_position_index = relative_coords.sum(-1)  # Wh*Ww, Wh*Ww

将其注册为buffer等待使用。

具体是作为索引，将relative_position_bias_table索引出来与attn相加。

shift window

不得不说，我在这里看了至少两个小时才弄懂这里的attn_mask究竟要干什么事

使用torch.roll对特征图进行位移。

Intuition: 因为要在windows之间传递信息，swin-T使用的是平移特征图，那么就会出现一个问题。假设是向左上方移（实际也是），原来图像上最左上方的像素会被循环平移到右下方，因此在右下方的window计算attention的时候，与距离过远的像素相互的attention不应该被考虑（应该被mask掉）

还记得在计算attention的时候传入的mask吗，它的作用是让具有相同index QK进行计算，而忽略不同index QK计算结果。

因此就有了下面这个我看了两个小时的代码：

这种广播机制我是真服气，太巧妙了

window_size = 2
input_resolution=(4,4)
shift_size = 1
if shift_size > 0:
    # calculate attention mask for SW-MSA
    H, W = input_resolution
    img_mask = torch.zeros((1, H, W, 1))  # 1 H W 1
    h_slices = (slice(0, -window_size),
                slice(-window_size, -shift_size),
                slice(-shift_size, None))
    w_slices = (slice(0, -window_size),
                slice(-window_size, -shift_size),
                slice(-shift_size, None))
    cnt = 0
    for h in h_slices:
        for w in w_slices:
            img_mask[:, h, w, :] = cnt
            cnt += 1
    mask_windows = window_partition(img_mask, window_size)  # nW, window_size, window_size, 1
    mask_windows = mask_windows.view(-1, window_size * window_size) # nW, window_size*window_size
    print('this is mask_windows')
    print(mask_windows)
    attn_mask = mask_windows.unsqueeze(1) - mask_windows.unsqueeze(2)
    print('-'*20)
    print('this is attn_mask')
    print(attn_mask)
    attn_mask = attn_mask.masked_fill(attn_mask != 0, float(-100.0)).masked_fill(attn_mask == 0, float(0.0))
else:
    attn_mask = None

输出：

一张图简述根据上述代码生成mask的方法

从上到下，从左到右，我们将window分别命名为w1, w2, w3, w4。在window中，由于每个像素都要与这个window中所有的像素计算attention，因此的窗中的attention是。

对于w1来说，所有的像素在平移之前是相邻的，因此不需要mask；对于w2而言，index为1和2的像素在原图中并不相邻，因此需要在attention中相应的地方进行mask；w3和w4以此类推。

而最核心的一点就是计算在attention中需要被mask掉的地方。
这里还请读者回忆如何计算attention，实际上就是q和k做矩阵乘法。

以w2为例
实际上这个顺序就是广播机制的顺序！

结语

至此，我们的Swin-T的所有代码都阅读完成啦。
之后我会继续和大家分享关于DL的论文和代码阅读。

Swin-T代码阅读作图.zip