import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.utils.checkpoint as checkpoint
import numpy as np
from typing import Optional
from thop import profile

class IRB(nn.Module):
    def __init__(self, in_features, hidden_features=None, out_features=None, ksize=3, act_layer=nn.Hardswish, drop=0.):
        super().__init__()
        out_features = out_features or in_features
        hidden_features = hidden_features or in_features
        self.fc1 = nn.Conv2d(in_features, hidden_features, 1, 1, 0)
        self.act = act_layer()
        self.conv = nn.Conv2d(hidden_features, hidden_features, kernel_size=ksize, padding=ksize // 2, stride=1,
                              groups=hidden_features)
        self.fc2 = nn.Conv2d(hidden_features, out_features, 1, 1, 0)
        self.drop = nn.Dropout(drop)

    def forward(self, x, H, W):
        B, N, C = x.shape
        x = x.permute(0, 2, 1).reshape(B, C, H, W)
        x = self.fc1(x)
        x = self.act(x)
        x = self.conv(x)
        x = self.act(x)
        x = self.fc2(x)
        return x.reshape(B, C, -1).permute(0, 2, 1)


def drop_path_f(x, drop_prob: float = 0., training: bool = False):

    if drop_prob == 0. or not training:
        return x
    keep_prob = 1 - drop_prob
    shape = (x.shape[0],) + (1,) * (x.ndim - 1)  # work with diff dim tensors, not just 2D ConvNets
    random_tensor = keep_prob + torch.rand(shape, dtype=x.dtype, device=x.device)
    random_tensor.floor_()  # binarize
    output = x.div(keep_prob) * random_tensor
    return output


class DropPath(nn.Module):

    def __init__(self, drop_prob=None):
        super(DropPath, self).__init__()
        self.drop_prob = drop_prob

    def forward(self, x):
        return drop_path_f(x, self.drop_prob, self.training)


def window_partition(x, window_size: int):
    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: int, H: int, W: int):

    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


class PatchEmbed2(nn.Module):
    def __init__(self, dim:int, patch_size=2, in_c=3, norm_layer=None):
        super().__init__()
        patch_size = (patch_size, patch_size)
        self.patch_size = patch_size
        self.in_chans = in_c
        self.embed_dim = dim
        self.proj = nn.Conv2d(dim, 2*dim, kernel_size=patch_size, stride=patch_size)
        self.norm = norm_layer(2*dim) if norm_layer else nn.Identity()

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

        x = x.view(B, H, W, C)
        pad_input = (H % self.patch_size[0] != 0) or (W % self.patch_size[1] != 0)
        if pad_input:
            x = F.pad(x, (0, self.patch_size[1] - W % self.patch_size[1],
                          0, self.patch_size[0] - H % self.patch_size[0],
                          0, 0))
        x = x.reshape(B, H, W, -1).permute(0, 3, 1, 2)
        x = self.proj(x)
        _, _, H, W = x.shape
        x = x.flatten(2).transpose(1, 2)
        x = self.norm(x)
        return x


class PatchEmbed(nn.Module):
    def __init__(self, patch_size=4, in_c=3, embed_dim=96, norm_layer=None):
        super().__init__()
        patch_size = (patch_size, patch_size)
        self.patch_size = patch_size
        self.in_chans = in_c
        self.embed_dim = embed_dim
        self.proj = nn.Conv2d(in_c, embed_dim, kernel_size=patch_size, stride=patch_size)
        self.norm = norm_layer(embed_dim) if norm_layer else nn.Identity()

    def forward(self, x):
        _, _, H, W = x.shape

        # padding
        # 如果输入图片的H，W不是patch_size的整数倍，需要进行padding
        pad_input = (H % self.patch_size[0] != 0) or (W % self.patch_size[1] != 0)
        if pad_input:
            x = F.pad(x, (0, self.patch_size[1] - W % self.patch_size[1],
                          0, self.patch_size[0] - H % self.patch_size[0],
                          0, 0))

        # 下采样patch_size倍
        x = self.proj(x)
        _, _, H, W = x.shape
        x = x.flatten(2).transpose(1, 2)
        x = self.norm(x)
        return x, H, W


class PatchMerging(nn.Module):

    def __init__(self, dim, norm_layer=nn.LayerNorm):
        super().__init__()
        self.dim = dim
        self.reduction = nn.Linear(4 * dim, 2 * dim, bias=False)
        self.norm = norm_layer(4 * dim)

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

        x = x.view(B, H, W, C)

        pad_input = (H % 2 == 1) or (W % 2 == 1)
        if pad_input:
            x = F.pad(x, (0, 0, 0, W % 2, 0, H % 2))

        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)  # [B, H/2*W/2, 2*C]

        return x


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.drop1 = nn.Dropout(drop)
        self.fc2 = nn.Linear(hidden_features, out_features)
        self.drop2 = nn.Dropout(drop)

    def forward(self, x):
        x = self.fc1(x)
        x = self.act(x)
        x = self.drop1(x)
        x = self.fc2(x)
        x = self.drop2(x)
        return x


class WindowAttention(nn.Module):

    def __init__(self, dim, window_size, num_heads, qkv_bias=True, attn_drop=0., proj_drop=0.):

        super().__init__()
        self.dim = dim
        self.window_size = window_size  # [Mh, Mw]
        self.num_heads = num_heads
        head_dim = dim // num_heads
        self.scale = 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*Mh-1 * 2*Mw-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]))
        coords_flatten = torch.flatten(coords, 1)  # [2, Mh*Mw]
        # [2, Mh*Mw, 1] - [2, 1, Mh*Mw]
        relative_coords = coords_flatten[:, :, None] - coords_flatten[:, None, :]  # [2, Mh*Mw, Mh*Mw]
        relative_coords = relative_coords.permute(1, 2, 0).contiguous()  # [Mh*Mw, Mh*Mw, 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)  # [Mh*Mw, Mh*Mw]
        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)

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

    def forward(self, x, mask: Optional[torch.Tensor] = 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)
        # [batch_size*num_windows, num_heads, Mh*Mw, embed_dim_per_head]
        q, k, v = qkv.unbind(0)  # make torchscript happy (cannot use tensor as tuple)

        # transpose: -> [batch_size*num_windows, num_heads, embed_dim_per_head, Mh*Mw]
        # @: multiply -> [batch_size*num_windows, num_heads, Mh*Mw, Mh*Mw]
        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)
        relative_position_bias = relative_position_bias.permute(2, 0, 1).contiguous()  # [nH, Mh*Mw, Mh*Mw]
        attn = attn + relative_position_bias.unsqueeze(0)

        if mask is not None:
            nW = mask.shape[0]  # num_windows
            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


class TransformerBlock(nn.Module):
    def __init__(self, dim, num_heads, window_sizes=(7,4,2), branch_num=3,
                 mlp_ratio=4., qkv_bias=True, drop=0., attn_drop=0., drop_path=0.,
                 act_layer=nn.GELU, norm_layer=nn.LayerNorm):
        super().__init__()
        self.dim = dim
        self.num_heads = num_heads
        self.window_sizes = window_sizes
        self.branch_num = branch_num
        self.mlp_ratio = mlp_ratio

        self.norm1 = norm_layer(dim)
        self.attn = WindowAttention(
            dim//branch_num, window_size=(self.window_sizes[0], self.window_sizes[0]), num_heads=num_heads//branch_num, qkv_bias=qkv_bias,
            attn_drop=attn_drop, proj_drop=drop)
        self.attn1 = WindowAttention(
            dim//branch_num, window_size=(self.window_sizes[1], self.window_sizes[1]), num_heads=num_heads//branch_num, qkv_bias=qkv_bias,
            attn_drop=attn_drop, proj_drop=drop)
        self.attn2 = WindowAttention(
            dim//branch_num, window_size=(self.window_sizes[2], self.window_sizes[2]), num_heads=num_heads//branch_num, qkv_bias=qkv_bias,
            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 = IRB(in_features=dim, hidden_features=mlp_hidden_dim, act_layer=act_layer, drop=drop)

    def forward(self, x, attn_mask):
        H, W = self.H, self.W
        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)
        x0 = x[:,:,:,:(C//self.branch_num)]
        x1 = x[:,:,:,(C//self.branch_num):(2*C//self.branch_num)]
        x2 = x[:,:,:,(2*C//self.branch_num):]
        # ----------------------------------------------------------------------------------------------
        pad_l = pad_t = 0
        pad_r = (self.window_sizes[0] - W % self.window_sizes[0]) % self.window_sizes[0]
        pad_b = (self.window_sizes[0] - H % self.window_sizes[0]) % self.window_sizes[0]
        x0 = F.pad(x0, (0, 0, pad_l, pad_r, pad_t, pad_b))
        _, Hp, Wp, _ = x0.shape
        attn_mask = None

        # partition windows
        x_windows = window_partition(x0, self.window_sizes[0])  # [nW*B, Mh, Mw, C]
        x_windows = x_windows.view(-1, self.window_sizes[0] * self.window_sizes[0], C//self.branch_num)  # [nW*B, Mh*Mw, C]

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

        # merge windows
        attn_windows = attn_windows.view(-1, self.window_sizes[0], self.window_sizes[0], C//self.branch_num)  # [nW*B, Mh, Mw, C]
        x0 = window_reverse(attn_windows, self.window_sizes[0], Hp, Wp)  # [B, H', W', C]

        if pad_r > 0 or pad_b > 0:
            # 把前面pad的数据移除掉
            x0 = x0[:, :H, :W, :].contiguous()

        x0 = x0.view(B, H * W, C//self.branch_num)
        # ----------------------------------------------------------------------------------------------
        pad_l = pad_t = 0
        pad_r = (self.window_sizes[1] - W % self.window_sizes[1]) % self.window_sizes[1]
        pad_b = (self.window_sizes[1] - H % self.window_sizes[1]) % self.window_sizes[1]
        x1 = F.pad(x1, (0, 0, pad_l, pad_r, pad_t, pad_b))
        _, Hp, Wp, _ = x1.shape
        attn_mask = None

        # partition windows
        x_windows = window_partition(x1, self.window_sizes[1])  # [nW*B, Mh, Mw, C]
        x_windows = x_windows.view(-1, self.window_sizes[1] * self.window_sizes[1],
                                   C // self.branch_num)  # [nW*B, Mh*Mw, C]

        # W-MSA/SW-MSA
        attn_windows = self.attn1(x_windows, mask=attn_mask)  # [nW*B, Mh*Mw, C]

        # merge windows
        attn_windows = attn_windows.view(-1, self.window_sizes[1], self.window_sizes[1],
                                         C // self.branch_num)  # [nW*B, Mh, Mw, C]
        x1 = window_reverse(attn_windows, self.window_sizes[1], Hp, Wp)  # [B, H', W', C]

        if pad_r > 0 or pad_b > 0:
            # 把前面pad的数据移除掉
            x1 = x1[:, :H, :W, :].contiguous()
        x1 = x1.view(B, H * W, C // self.branch_num)
        # ----------------------------------------------------------------------------------------------
        pad_l = pad_t = 0
        pad_r = (self.window_sizes[2] - W % self.window_sizes[2]) % self.window_sizes[2]
        pad_b = (self.window_sizes[2] - H % self.window_sizes[2]) % self.window_sizes[2]
        x2 = F.pad(x2, (0, 0, pad_l, pad_r, pad_t, pad_b))
        _, Hp, Wp, _ = x2.shape
        attn_mask = None
        x_windows = window_partition(x2, self.window_sizes[2])  # [nW*B, Mh, Mw, C]
        x_windows = x_windows.view(-1, self.window_sizes[2] * self.window_sizes[2],
                                   C // self.branch_num)  # [nW*B, Mh*Mw, C]

        attn_windows = self.attn2(x_windows, mask=attn_mask)  # [nW*B, Mh*Mw, C]

        attn_windows = attn_windows.view(-1, self.window_sizes[2], self.window_sizes[2],
                                         C // self.branch_num)  # [nW*B, Mh, Mw, C]
        x2 = window_reverse(attn_windows, self.window_sizes[2], Hp, Wp)  # [B, H', W', C]

        if pad_r > 0 or pad_b > 0:
            x2 = x2[:, :H, :W, :].contiguous()

        x2 = x2.view(B, H * W, C // self.branch_num)

        x = torch.cat([x0, x1, x2], -1)
        # FFN
        x = shortcut + self.drop_path(x)
        x = x + self.drop_path(self.mlp(self.norm2(x), H, W))

        return x


class BasicLayer(nn.Module):
    def __init__(self, dim, depth, num_heads, window_size,
                 mlp_ratio=4., qkv_bias=True, drop=0., attn_drop=0.,
                 drop_path=0., norm_layer=nn.LayerNorm, downsample=None, use_checkpoint=False):
        super().__init__()
        self.dim = dim
        self.depth = depth
        self.window_size = window_size
        self.use_checkpoint = use_checkpoint
        self.shift_size = window_size // 2

        # build blocks
        self.blocks = nn.ModuleList([
            TransformerBlock(
                dim=dim,
                num_heads=num_heads,
                window_sizes=(7,4,2),
                mlp_ratio=mlp_ratio,
                qkv_bias=qkv_bias,
                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(dim=dim, norm_layer=norm_layer)
        else:
            self.downsample = None

    def create_mask(self, x, H, W):
        Hp = int(np.ceil(H / self.window_size)) * self.window_size
        Wp = int(np.ceil(W / self.window_size)) * self.window_size
        img_mask = torch.zeros((1, Hp, Wp, 1), device=x.device)  # [1, Hp, Wp, 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, Mh, Mw, 1]
        mask_windows = mask_windows.view(-1, self.window_size * self.window_size)  # [nW, Mh*Mw]
        attn_mask = mask_windows.unsqueeze(1) - mask_windows.unsqueeze(2)  # [nW, 1, Mh*Mw] - [nW, Mh*Mw, 1]
        attn_mask = attn_mask.masked_fill(attn_mask != 0, float(-100.0)).masked_fill(attn_mask == 0, float(0.0))
        return attn_mask

    def forward(self, x, H, W):
        attn_mask = self.create_mask(x, H, W)  # [nW, Mh*Mw, Mh*Mw]
        for blk in self.blocks:
            blk.H, blk.W = H, W
            if not torch.jit.is_scripting() and self.use_checkpoint:
                x = checkpoint.checkpoint(blk, x, attn_mask)
            else:
                x = blk(x, attn_mask)
        if self.downsample is not None:
            x = self.downsample(x, H, W)
            H, W = (H + 1) // 2, (W + 1) // 2

        return x, H, W


class Transformer(nn.Module):
    def __init__(self, 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,
                 drop_rate=0., attn_drop_rate=0., drop_path_rate=0.1,
                 norm_layer=nn.LayerNorm, 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.patch_norm = patch_norm
        # stage4输出特征矩阵的channels
        self.num_features = int(embed_dim * 2 ** (self.num_layers - 1))
        self.mlp_ratio = mlp_ratio

        self.patch_embed = PatchEmbed(
            patch_size=patch_size, in_c=in_chans, embed_dim=embed_dim,
            norm_layer=norm_layer if self.patch_norm else None)
        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

        self.layers = nn.ModuleList()
        for i_layer in range(self.num_layers):
            layers = BasicLayer(dim=int(embed_dim * 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,
                                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=PatchEmbed2 if (i_layer < self.num_layers - 1) else None,
                                use_checkpoint=use_checkpoint)
            self.layers.append(layers)

        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):
            nn.init.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(self, x):
        # x: [B, L, C]
        x, H, W = self.patch_embed(x)
        x = self.pos_drop(x)

        for layer in self.layers:
            x, H, W = layer(x, H, W)

        x = self.norm(x)  # [B, L, C]
        x = self.avgpool(x.transpose(1, 2))  # [B, C, 1]
        x = torch.flatten(x, 1)
        x = self.head(x)
        return x


def MWT(num_classes: int = 1000, **kwargs):
    model = Transformer(in_chans=3,
                        patch_size=4,
                        # window_sizes=(7,4,2),
                        embed_dim=96,
                        depths=(2, 4, 4, 2),
                        num_heads=(3, 6, 12, 24),
                        num_classes=num_classes,
                        **kwargs)
    return model

if __name__ == '__main__':
    model = MWT(num_classes=2)
    input = torch.randn(1, 3, 224, 224)
    flops, params = profile(model, inputs=(input,))
    print(flops)
    print(params)