Swin—UNet代码学习（pytorch）

 model流程学习： 

self.swin_unet = SwinTransformerSys(img_size=config.DATA.IMG_SIZE,
                                patch_size=config.MODEL.SWIN.PATCH_SIZE,
                                in_chans=config.MODEL.SWIN.IN_CHANS,
                                num_classes=self.num_classes,
                                embed_dim=config.MODEL.SWIN.EMBED_DIM,
                                depths=config.MODEL.SWIN.DEPTHS,
                                num_heads=config.MODEL.SWIN.NUM_HEADS,
                                window_size=config.MODEL.SWIN.WINDOW_SIZE,
                                mlp_ratio=config.MODEL.SWIN.MLP_RATIO,
                                qkv_bias=config.MODEL.SWIN.QKV_BIAS,
                                qk_scale=config.MODEL.SWIN.QK_SCALE,
                                drop_rate=config.MODEL.DROP_RATE,
                                drop_path_rate=config.MODEL.DROP_PATH_RATE,
                                ape=config.MODEL.SWIN.APE,
                                patch_norm=config.MODEL.SWIN.PATCH_NORM,
                                use_checkpoint=config.TRAIN.USE_CHECKPOINT)

图片大小设置为224*224*3（输入图片为3通道的），patch_size:4，num_class:21843(类别数量)，embed_dim：线性投影输出通道数

Swin Transformer

参数设置

def __init__(self, img_size=224, patch_size=4, in_chans=3, num_classes=1000,
                 embed_dim=96, depths=[2, 2, 2, 2], depths_decoder=[1, 2, 2, 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, final_upsample="expand_first", **kwargs):

Patch Embedding

划分patch：模型先将图像分割成每块大小为 4x4的patch，那么就会有56x56个patch。--patchEmbed。

编码初始特征：conv卷积（3，96，4，4），经过norm后56x56x96。

# 将图片分割为不重叠的面片
        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

Attention:预处理完成后，维度是H/4XW/4XC。

BasicLayer：

PatchMerging：downsample

 每个block是W-MSA（window-multihead self attention）或者SW-MSA（shift window multihead self attention），一般有偶数个block，两种block交替出现，比如6个block，0，2，4是W-MSA，1，3，5是SW-MSA。在经历完一个stage后，会进行下采样，依旧是56*56

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

个patch，四个为一组，分别取每组中的左上，右上、左下、右下堆叠一起，经过一个layernorm，linear层，实现维度下采样、特征加倍的效果。实际上它可以看成一种加权池化的过程.--进行四次下采样。

 SwinTransformerBlock：

shift_size<windowsize<imput_resolution

windowattention:

class WindowAttention(nn.Module):
    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

        # 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 = q * self.scale
        attn = (q @ k.transpose(-2, -1))

        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

参考：Swin-Transformer结合代码深度解析 - 知乎 (zhihu.com)

个人学习笔记（主要自用）。如有不对，敬请指正！