# pytorch_diffusion + derived encoder decoder
import math
import torch
import torch.nn as nn
import numpy as np


def get_timestep_embedding(timesteps, embedding_dim):
    """
    This matches the implementation in Denoising Diffusion Probabilistic Models:
    From Fairseq.
    Build sinusoidal embeddings.
    This matches the implementation in tensor2tensor, but differs slightly
    from the description in Section 3.5 of "Attention Is All You Need".
    """
    assert len(timesteps.shape) == 1

    half_dim = embedding_dim // 2
    emb = math.log(10000) / (half_dim - 1)
    emb = torch.exp(torch.arange(half_dim, dtype=torch.float32) * -emb)
    emb = emb.to(device=timesteps.device)
    emb = timesteps.float()[:, None] * emb[None, :]
    emb = torch.cat([torch.sin(emb), torch.cos(emb)], dim=1)
    if embedding_dim % 2 == 1:  # zero pad
        emb = torch.nn.functional.pad(emb, (0, 1, 0, 0))
    return emb


def nonlinearity(x):
    # swish
    return x * torch.sigmoid(x)


class SpatialNorm(nn.Module):
    def __init__(
        self,
        f_channels,
        zq_channels,
        norm_layer=nn.GroupNorm,
        freeze_norm_layer=False,
        add_conv=False,
        **norm_layer_params,
    ):
        super().__init__()
        self.norm_layer = norm_layer(num_channels=f_channels, **norm_layer_params)
        if freeze_norm_layer:
            for p in self.norm_layer.parameters:
                p.requires_grad = False
        self.add_conv = add_conv
        if self.add_conv:
            self.conv = nn.Conv2d(zq_channels, zq_channels, kernel_size=3, stride=1, padding=1)
        self.conv_y = nn.Conv2d(zq_channels, f_channels, kernel_size=1, stride=1, padding=0)
        self.conv_b = nn.Conv2d(zq_channels, f_channels, kernel_size=1, stride=1, padding=0)

    def forward(self, f, zq):
        f_size = f.shape[-2:]
        zq = torch.nn.functional.interpolate(zq, size=f_size, mode="nearest")
        if self.add_conv:
            zq = self.conv(zq)
        norm_f = self.norm_layer(f)
        new_f = norm_f * self.conv_y(zq) + self.conv_b(zq)
        return new_f


def Normalize(in_channels, zq_ch, add_conv):
    return SpatialNorm(
        in_channels,
        zq_ch,
        norm_layer=nn.GroupNorm,
        freeze_norm_layer=False,
        add_conv=add_conv,
        num_groups=32,
        eps=1e-6,
        affine=True,
    )


class Upsample(nn.Module):
    def __init__(self, in_channels, with_conv):
        super().__init__()
        self.with_conv = with_conv
        if self.with_conv:
            self.conv = torch.nn.Conv2d(
                in_channels, in_channels, kernel_size=3, stride=1, padding=1
            )

    def forward(self, x):
        x = torch.nn.functional.interpolate(x, scale_factor=2.0, mode="nearest")
        if self.with_conv:
            x = self.conv(x)
        return x


class Downsample(nn.Module):
    def __init__(self, in_channels, with_conv):
        super().__init__()
        self.with_conv = with_conv
        if self.with_conv:
            # no asymmetric padding in torch conv, must do it ourselves
            self.conv = torch.nn.Conv2d(
                in_channels, in_channels, kernel_size=3, stride=2, padding=0
            )

    def forward(self, x):
        if self.with_conv:
            pad = (0, 1, 0, 1)
            x = torch.nn.functional.pad(x, pad, mode="constant", value=0)
            x = self.conv(x)
        else:
            x = torch.nn.functional.avg_pool2d(x, kernel_size=2, stride=2)
        return x


class ResnetBlock(nn.Module):
    def __init__(
        self,
        *,
        in_channels,
        out_channels=None,
        conv_shortcut=False,
        dropout,
        temb_channels=512,
        zq_ch=None,
        add_conv=False,
    ):
        super().__init__()
        self.in_channels = in_channels
        out_channels = in_channels if out_channels is None else out_channels
        self.out_channels = out_channels
        self.use_conv_shortcut = conv_shortcut

        self.norm1 = Normalize(in_channels, zq_ch, add_conv=add_conv)
        self.conv1 = torch.nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=1, padding=1)
        if temb_channels > 0:
            self.temb_proj = torch.nn.Linear(temb_channels, out_channels)
        self.norm2 = Normalize(out_channels, zq_ch, add_conv=add_conv)
        self.dropout = torch.nn.Dropout(dropout)
        self.conv2 = torch.nn.Conv2d(out_channels, out_channels, kernel_size=3, stride=1, padding=1)
        if self.in_channels != self.out_channels:
            if self.use_conv_shortcut:
                self.conv_shortcut = torch.nn.Conv2d(
                    in_channels, out_channels, kernel_size=3, stride=1, padding=1
                )
            else:
                self.nin_shortcut = torch.nn.Conv2d(
                    in_channels, out_channels, kernel_size=1, stride=1, padding=0
                )

    def forward(self, x, temb, zq):
        h = x
        h = self.norm1(h, zq)
        h = nonlinearity(h)
        h = self.conv1(h)

        if temb is not None:
            h = h + self.temb_proj(nonlinearity(temb))[:, :, None, None]

        h = self.norm2(h, zq)
        h = nonlinearity(h)
        h = self.dropout(h)
        h = self.conv2(h)

        if self.in_channels != self.out_channels:
            if self.use_conv_shortcut:
                x = self.conv_shortcut(x)
            else:
                x = self.nin_shortcut(x)

        return x + h


class AttnBlock(nn.Module):
    def __init__(self, in_channels, zq_ch=None, add_conv=False):
        super().__init__()
        self.in_channels = in_channels

        self.norm = Normalize(in_channels, zq_ch, add_conv=add_conv)
        self.q = torch.nn.Conv2d(in_channels, in_channels, kernel_size=1, stride=1, padding=0)
        self.k = torch.nn.Conv2d(in_channels, in_channels, kernel_size=1, stride=1, padding=0)
        self.v = torch.nn.Conv2d(in_channels, in_channels, kernel_size=1, stride=1, padding=0)
        self.proj_out = torch.nn.Conv2d(
            in_channels, in_channels, kernel_size=1, stride=1, padding=0
        )

    def forward(self, x, zq):
        h_ = x
        h_ = self.norm(h_, zq)
        q = self.q(h_)
        k = self.k(h_)
        v = self.v(h_)

        # compute attention
        b, c, h, w = q.shape
        q = q.reshape(b, c, h * w)
        q = q.permute(0, 2, 1)  # b,hw,c
        k = k.reshape(b, c, h * w)  # b,c,hw
        w_ = torch.bmm(q, k)  # b,hw,hw    w[b,i,j]=sum_c q[b,i,c]k[b,c,j]
        w_ = w_ * (int(c) ** (-0.5))
        w_ = torch.nn.functional.softmax(w_, dim=2)

        # attend to values
        v = v.reshape(b, c, h * w)
        w_ = w_.permute(0, 2, 1)  # b,hw,hw (first hw of k, second of q)
        h_ = torch.bmm(v, w_)  # b, c,hw (hw of q) h_[b,c,j] = sum_i v[b,c,i] w_[b,i,j]
        h_ = h_.reshape(b, c, h, w)

        h_ = self.proj_out(h_)

        return x + h_


class MOVQDecoder(nn.Module):
    def __init__(
        self,
        *,
        ch,
        out_ch,
        ch_mult=(1, 2, 4, 8),
        num_res_blocks,
        attn_resolutions,
        dropout=0.0,
        resamp_with_conv=True,
        in_channels,
        resolution,
        z_channels,
        give_pre_end=False,
        zq_ch=None,
        add_conv=False,
        **ignorekwargs,
    ):
        super().__init__()
        self.ch = ch
        self.temb_ch = 0
        self.num_resolutions = len(ch_mult)
        self.num_res_blocks = num_res_blocks
        self.resolution = resolution
        self.in_channels = in_channels
        self.give_pre_end = give_pre_end

        # compute in_ch_mult, block_in and curr_res at lowest res
        in_ch_mult = (1,) + tuple(ch_mult)
        block_in = ch * ch_mult[self.num_resolutions - 1]
        curr_res = resolution // 2 ** (self.num_resolutions - 1)
        self.z_shape = (1, z_channels, curr_res, curr_res)
        print(
            "Working with z of shape {} = {} dimensions.".format(
                self.z_shape, np.prod(self.z_shape)
            )
        )

        # z to block_in
        self.conv_in = torch.nn.Conv2d(z_channels, block_in, kernel_size=3, stride=1, padding=1)

        # middle
        self.mid = nn.Module()
        self.mid.block_1 = ResnetBlock(
            in_channels=block_in,
            out_channels=block_in,
            temb_channels=self.temb_ch,
            dropout=dropout,
            zq_ch=zq_ch,
            add_conv=add_conv,
        )
        self.mid.attn_1 = AttnBlock(block_in, zq_ch, add_conv=add_conv)
        self.mid.block_2 = ResnetBlock(
            in_channels=block_in,
            out_channels=block_in,
            temb_channels=self.temb_ch,
            dropout=dropout,
            zq_ch=zq_ch,
            add_conv=add_conv,
        )

        # upsampling
        self.up = nn.ModuleList()
        for i_level in reversed(range(self.num_resolutions)):
            block = nn.ModuleList()
            attn = nn.ModuleList()
            block_out = ch * ch_mult[i_level]
            for i_block in range(self.num_res_blocks + 1):
                block.append(
                    ResnetBlock(
                        in_channels=block_in,
                        out_channels=block_out,
                        temb_channels=self.temb_ch,
                        dropout=dropout,
                        zq_ch=zq_ch,
                        add_conv=add_conv,
                    )
                )
                block_in = block_out
                if curr_res in attn_resolutions:
                    attn.append(AttnBlock(block_in, zq_ch, add_conv=add_conv))
            up = nn.Module()
            up.block = block
            up.attn = attn
            if i_level != 0:
                up.upsample = Upsample(block_in, resamp_with_conv)
                curr_res = curr_res * 2
            self.up.insert(0, up)  # prepend to get consistent order

        # end
        self.norm_out = Normalize(block_in, zq_ch, add_conv=add_conv)
        self.conv_out = torch.nn.Conv2d(block_in, out_ch, kernel_size=3, stride=1, padding=1)

    def forward(self, z, zq):
        # assert z.shape[1:] == self.z_shape[1:]
        self.last_z_shape = z.shape

        # timestep embedding
        temb = None

        # z to block_in
        h = self.conv_in(z)

        # middle
        h = self.mid.block_1(h, temb, zq)
        h = self.mid.attn_1(h, zq)
        h = self.mid.block_2(h, temb, zq)

        # upsampling
        for i_level in reversed(range(self.num_resolutions)):
            for i_block in range(self.num_res_blocks + 1):
                h = self.up[i_level].block[i_block](h, temb, zq)
                if len(self.up[i_level].attn) > 0:
                    h = self.up[i_level].attn[i_block](h, zq)
            if i_level != 0:
                h = self.up[i_level].upsample(h)

        # end
        if self.give_pre_end:
            return h

        h = self.norm_out(h, zq)
        h = nonlinearity(h)
        h = self.conv_out(h)
        return h

    def forward_with_features_output(self, z, zq):
        # assert z.shape[1:] == self.z_shape[1:]
        self.last_z_shape = z.shape

        # timestep embedding
        temb = None
        output_features = {}

        # z to block_in
        h = self.conv_in(z)
        output_features["conv_in"] = h

        # middle
        h = self.mid.block_1(h, temb, zq)
        output_features["mid_block_1"] = h
        h = self.mid.attn_1(h, zq)
        output_features["mid_attn_1"] = h
        h = self.mid.block_2(h, temb, zq)
        output_features["mid_block_2"] = h

        # upsampling
        for i_level in reversed(range(self.num_resolutions)):
            for i_block in range(self.num_res_blocks + 1):
                h = self.up[i_level].block[i_block](h, temb, zq)
                output_features[f"up_{i_level}_block_{i_block}"] = h
                if len(self.up[i_level].attn) > 0:
                    h = self.up[i_level].attn[i_block](h, zq)
                    output_features[f"up_{i_level}_attn_{i_block}"] = h
            if i_level != 0:
                h = self.up[i_level].upsample(h)
                output_features[f"up_{i_level}_upsample"] = h

        # end
        if self.give_pre_end:
            return h

        h = self.norm_out(h, zq)
        output_features["norm_out"] = h
        h = nonlinearity(h)
        output_features["nonlinearity"] = h
        h = self.conv_out(h)
        output_features["conv_out"] = h

        return h, output_features