# Copyright (c) Alibaba, Inc. and its affiliates.
"""
This model is taken from
https://github.com/SamsungLabs/EdgeViTs
"""

from collections import OrderedDict
from functools import partial

import torch
import torch.nn as nn
from timm.models.layers import DropPath, to_2tuple, trunc_normal_

from easycv.models.utils import ConvMlp, Mlp
from easycv.utils.checkpoint import load_checkpoint
from easycv.utils.logger import get_root_logger
from ..registry import BACKBONES


class GlobalSparseAttn(nn.Module):

    def __init__(self,
                 dim,
                 num_heads=8,
                 qkv_bias=False,
                 qk_scale=None,
                 attn_drop=0.,
                 proj_drop=0.,
                 sr_ratio=1):
        super().__init__()
        self.num_heads = num_heads
        head_dim = dim // num_heads
        # NOTE scale factor was wrong in my original version, can set manually to be compat with prev weights
        self.scale = qk_scale or head_dim**-0.5

        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)

        # self.upsample = nn.Upsample(scale_factor=sr_ratio, mode='nearest')
        self.sr = sr_ratio
        if self.sr > 1:
            self.sampler = nn.AvgPool2d(1, sr_ratio)
            kernel_size = sr_ratio
            self.LocalProp = nn.ConvTranspose2d(
                dim, dim, kernel_size, stride=sr_ratio, groups=dim)
            self.norm = nn.LayerNorm(dim)
        else:
            self.sampler = nn.Identity()
            self.upsample = nn.Identity()
            self.norm = nn.Identity()

    def forward(self, x, H: int, W: int):
        B, N, C = x.shape
        if self.sr > 1.:
            x = x.transpose(1, 2).reshape(B, C, H, W)
            x = self.sampler(x)
            x = x.flatten(2).transpose(1, 2)

        qkv = self.qkv(x).reshape(B, -1, 3, self.num_heads,
                                  C // self.num_heads).permute(2, 0, 3, 1, 4)
        q, k, v = qkv[0], qkv[1], qkv[2]

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

        x = (attn @ v).transpose(1, 2).reshape(B, -1, C)

        if self.sr > 1:
            x = x.permute(0, 2, 1).reshape(B, C, int(H / self.sr),
                                           int(W / self.sr))
            x = self.LocalProp(x)
            x = x.reshape(B, C, -1).permute(0, 2, 1)
            x = self.norm(x)

        x = self.proj(x)
        x = self.proj_drop(x)
        return x


class LocalAgg(nn.Module):

    def __init__(self,
                 dim,
                 num_heads,
                 mlp_ratio=4.,
                 qkv_bias=False,
                 qk_scale=None,
                 drop=0.,
                 attn_drop=0.,
                 drop_path=0.,
                 act_layer=nn.GELU,
                 norm_layer=nn.LayerNorm):
        super().__init__()
        self.pos_embed = nn.Conv2d(dim, dim, 3, padding=1, groups=dim)
        self.norm1 = nn.BatchNorm2d(dim)
        self.conv1 = nn.Conv2d(dim, dim, 1)
        self.conv2 = nn.Conv2d(dim, dim, 1)
        self.attn = nn.Conv2d(dim, dim, 5, padding=2, groups=dim)
        self.drop_path = DropPath(
            drop_path) if drop_path > 0. else nn.Identity()
        self.norm2 = nn.BatchNorm2d(dim)
        mlp_hidden_dim = int(dim * mlp_ratio)
        self.mlp = ConvMlp(
            in_features=dim,
            hidden_features=mlp_hidden_dim,
            act_layer=act_layer,
            drop=drop)

    def forward(self, x):
        x = x + self.pos_embed(x)
        x = x + self.drop_path(
            self.conv2(self.attn(self.conv1(self.norm1(x)))))
        x = x + self.drop_path(self.mlp(self.norm2(x)))
        return x


class SelfAttn(nn.Module):

    def __init__(self,
                 dim,
                 num_heads,
                 mlp_ratio=4.,
                 qkv_bias=False,
                 qk_scale=None,
                 drop=0.,
                 attn_drop=0.,
                 drop_path=0.,
                 act_layer=nn.GELU,
                 norm_layer=nn.LayerNorm,
                 sr_ratio=1.):
        super().__init__()
        self.pos_embed = nn.Conv2d(dim, dim, 3, padding=1, groups=dim)
        self.norm1 = norm_layer(dim)
        self.attn = GlobalSparseAttn(
            dim,
            num_heads=num_heads,
            qkv_bias=qkv_bias,
            qk_scale=qk_scale,
            attn_drop=attn_drop,
            proj_drop=drop,
            sr_ratio=sr_ratio)
        # NOTE: drop path for stochastic depth, we shall see if this is better than dropout here
        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)
        # global layer_scale
        # self.ls = layer_scale

    def forward(self, x):
        x = x + self.pos_embed(x)
        B, N, H, W = x.shape
        x = x.flatten(2).transpose(1, 2)
        x = x + self.drop_path(self.attn(self.norm1(x), H, W))
        x = x + self.drop_path(self.mlp(self.norm2(x)))
        x = x.transpose(1, 2).reshape(B, N, H, W)
        return x


class LGLBlock(nn.Module):

    def __init__(self,
                 dim,
                 num_heads,
                 mlp_ratio=4.,
                 qkv_bias=False,
                 qk_scale=None,
                 drop=0.,
                 attn_drop=0.,
                 drop_path=0.,
                 act_layer=nn.GELU,
                 norm_layer=nn.LayerNorm,
                 sr_ratio=1.):
        super().__init__()

        if sr_ratio > 1:
            self.LocalAgg = LocalAgg(dim, num_heads, mlp_ratio, qkv_bias,
                                     qk_scale, drop, attn_drop, drop_path,
                                     act_layer, norm_layer)
        else:
            self.LocalAgg = nn.Identity()

        self.SelfAttn = SelfAttn(dim, num_heads, mlp_ratio, qkv_bias, qk_scale,
                                 drop, attn_drop, drop_path, act_layer,
                                 norm_layer, sr_ratio)

    def forward(self, x):
        x = self.LocalAgg(x)
        x = self.SelfAttn(x)
        return x


class PatchEmbed(nn.Module):
    """ Image to Patch Embedding
    """

    def __init__(self, img_size=224, patch_size=16, in_chans=3, embed_dim=768):
        super().__init__()
        img_size = to_2tuple(img_size)
        patch_size = to_2tuple(patch_size)
        num_patches = (img_size[1] // patch_size[1]) * (
            img_size[0] // patch_size[0])
        self.img_size = img_size
        self.patch_size = patch_size
        self.num_patches = num_patches
        self.norm = nn.LayerNorm(embed_dim)
        self.proj = nn.Conv2d(
            in_chans, embed_dim, kernel_size=patch_size, stride=patch_size)

    def forward(self, x):
        B, C, H, W = x.shape
        assert H == self.img_size[0] and W == self.img_size[1], \
            f"Input image size ({B}*{C}*{H}*{W}) doesn't match model ({self.img_size[0]}*{self.img_size[1]})."
        x = self.proj(x)
        B, C, H, W = x.shape
        x = x.flatten(2).transpose(1, 2)
        x = self.norm(x)
        x = x.reshape(B, H, W, -1).permute(0, 3, 1, 2).contiguous()
        return x


@BACKBONES.register_module()
class EdgeVit(nn.Module):
    """ Vision Transformer
    A PyTorch impl of : `An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale`  -
        https://arxiv.org/abs/2010.11929
    """

    def __init__(self,
                 depth=[1, 2, 3, 2],
                 img_size=224,
                 in_chans=3,
                 num_classes=1000,
                 embed_dim=[48, 96, 240, 384],
                 head_dim=48,
                 mlp_ratio=[4] * 4,
                 qkv_bias=True,
                 qk_scale=None,
                 representation_size=None,
                 drop_rate=0.,
                 attn_drop_rate=0.,
                 drop_path_rate=0.,
                 norm_layer=partial(nn.LayerNorm, eps=1e-8),
                 sr_ratios=[4, 2, 2, 1],
                 pretrained=None):
        """
        Args:
            depth (list): depth of each stage
            img_size (int, tuple): input image size
            in_chans (int): number of input channels
            num_classes (int): number of classes for classification head
            embed_dim (list): embedding dimension of each stage
            head_dim (int): head dimension
            mlp_ratio (int): ratio of mlp hidden dim to embedding dim
            qkv_bias (bool): enable bias for qkv if True
            qk_scale (float): override default qk scale of head_dim ** -0.5 if set
            representation_size (Optional[int]): enable and set representation layer (pre-logits) to this value if set
            drop_rate (float): dropout rate
            attn_drop_rate (float): attention dropout rate
            drop_path_rate (float): stochastic depth rate
            norm_layer (nn.Module): normalization layer
        """
        super().__init__()
        self.num_classes = num_classes
        self.num_features = self.embed_dim = embed_dim  # num_features for consistency with other models
        norm_layer = norm_layer or partial(nn.LayerNorm, eps=1e-6)

        self.patch_embed1 = PatchEmbed(
            img_size=img_size,
            patch_size=4,
            in_chans=in_chans,
            embed_dim=embed_dim[0])
        self.patch_embed2 = PatchEmbed(
            img_size=img_size // 4,
            patch_size=2,
            in_chans=embed_dim[0],
            embed_dim=embed_dim[1])
        self.patch_embed3 = PatchEmbed(
            img_size=img_size // 8,
            patch_size=2,
            in_chans=embed_dim[1],
            embed_dim=embed_dim[2])
        self.patch_embed4 = PatchEmbed(
            img_size=img_size // 16,
            patch_size=2,
            in_chans=embed_dim[2],
            embed_dim=embed_dim[3])

        self.pos_drop = nn.Dropout(p=drop_rate)
        dpr = [
            x.item() for x in torch.linspace(0, drop_path_rate, sum(depth))
        ]  # stochastic depth decay rule
        num_heads = [dim // head_dim for dim in embed_dim]
        self.blocks1 = nn.ModuleList([
            LGLBlock(
                dim=embed_dim[0],
                num_heads=num_heads[0],
                mlp_ratio=mlp_ratio[0],
                qkv_bias=qkv_bias,
                qk_scale=qk_scale,
                drop=drop_rate,
                attn_drop=attn_drop_rate,
                drop_path=dpr[i],
                norm_layer=norm_layer,
                sr_ratio=sr_ratios[0]) for i in range(depth[0])
        ])
        self.blocks2 = nn.ModuleList([
            LGLBlock(
                dim=embed_dim[1],
                num_heads=num_heads[1],
                mlp_ratio=mlp_ratio[1],
                qkv_bias=qkv_bias,
                qk_scale=qk_scale,
                drop=drop_rate,
                attn_drop=attn_drop_rate,
                drop_path=dpr[i + depth[0]],
                norm_layer=norm_layer,
                sr_ratio=sr_ratios[1]) for i in range(depth[1])
        ])
        self.blocks3 = nn.ModuleList([
            LGLBlock(
                dim=embed_dim[2],
                num_heads=num_heads[2],
                mlp_ratio=mlp_ratio[2],
                qkv_bias=qkv_bias,
                qk_scale=qk_scale,
                drop=drop_rate,
                attn_drop=attn_drop_rate,
                drop_path=dpr[i + depth[0] + depth[1]],
                norm_layer=norm_layer,
                sr_ratio=sr_ratios[2]) for i in range(depth[2])
        ])
        self.blocks4 = nn.ModuleList([
            LGLBlock(
                dim=embed_dim[3],
                num_heads=num_heads[3],
                mlp_ratio=mlp_ratio[3],
                qkv_bias=qkv_bias,
                qk_scale=qk_scale,
                drop=drop_rate,
                attn_drop=attn_drop_rate,
                drop_path=dpr[i + depth[0] + depth[1] + depth[2]],
                norm_layer=norm_layer,
                sr_ratio=sr_ratios[3]) for i in range(depth[3])
        ])
        self.norm = nn.BatchNorm2d(embed_dim[-1])

        # Representation layer
        if representation_size:
            self.num_features = representation_size
            self.pre_logits = nn.Sequential(
                OrderedDict([('fc', nn.Linear(embed_dim, representation_size)),
                             ('act', nn.Tanh())]))
        else:
            self.pre_logits = nn.Identity()

        self.pretrained = pretrained
        self.init_weights()

    def init_weights(self, pretrained=None):
        """Initialize the weights in backbone.
        Args:
            pretrained (str, optional): Path to pre-trained weights.
                Defaults to None.
        """
        pretrained = pretrained or self.pretrained

        def _init_weights(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)

        if isinstance(pretrained, str):
            self.apply(_init_weights)
            logger = get_root_logger()
            load_checkpoint(self, pretrained, strict=False, logger=logger)
        elif pretrained is None:
            self.apply(_init_weights)
        else:
            raise TypeError('pretrained must be a str or None')

    @torch.jit.ignore
    def no_weight_decay(self):
        return {'pos_embed', 'cls_token'}

    def forward_features(self, x):
        x = self.patch_embed1(x)
        x = self.pos_drop(x)
        for blk in self.blocks1:
            x = blk(x)
        x = self.patch_embed2(x)
        for blk in self.blocks2:
            x = blk(x)
        x = self.patch_embed3(x)
        for blk in self.blocks3:
            x = blk(x)
        x = self.patch_embed4(x)
        for blk in self.blocks4:
            x = blk(x)
        x = self.norm(x)
        x = self.pre_logits(x)
        return x

    def forward(self, x):
        x = self.forward_features(x)
        return [x]