imagedream/ldm/modules/attention.py

from inspect import isfunction
import math
import torch
import torch.nn.functional as F
from torch import nn, einsum
from einops import rearrange, repeat
from typing import Optional, Any

from .diffusionmodules.util import checkpoint


try:
    import xformers
    import xformers.ops
    HAS_XFORMERS = True
except ImportError:
    HAS_XFORMERS = False

# CrossAttn precision handling
import os

_ATTN_PRECISION = os.environ.get("ATTN_PRECISION", "fp32")


def exists(val):
    return val is not None


def uniq(arr):
    return {el: True for el in arr}.keys()


def default(val, d):
    if exists(val):
        return val
    return d() if isfunction(d) else d


def max_neg_value(t):
    return -torch.finfo(t.dtype).max


def init_(tensor):
    dim = tensor.shape[-1]
    std = 1 / math.sqrt(dim)
    tensor.uniform_(-std, std)
    return tensor


# feedforward
class GEGLU(nn.Module):
    def __init__(self, dim_in, dim_out):
        super().__init__()
        self.proj = nn.Linear(dim_in, dim_out * 2)

    def forward(self, x):
        x, gate = self.proj(x).chunk(2, dim=-1)
        return x * F.gelu(gate)


class FeedForward(nn.Module):
    def __init__(self, dim, dim_out=None, mult=4, glu=False, dropout=0.0):
        super().__init__()
        inner_dim = int(dim * mult)
        dim_out = default(dim_out, dim)
        project_in = (
            nn.Sequential(nn.Linear(dim, inner_dim), nn.GELU())
            if not glu
            else GEGLU(dim, inner_dim)
        )

        self.net = nn.Sequential(
            project_in, nn.Dropout(dropout), nn.Linear(inner_dim, dim_out)
        )

    def forward(self, x):
        return self.net(x)


def zero_module(module):
    """
    Zero out the parameters of a module and return it.
    """
    for p in module.parameters():
        p.detach().zero_()
    return module


def Normalize(in_channels):
    return torch.nn.GroupNorm(
        num_groups=32, num_channels=in_channels, eps=1e-6, affine=True
    )


class SpatialSelfAttention(nn.Module):
    def __init__(self, in_channels):
        super().__init__()
        self.in_channels = in_channels

        self.norm = Normalize(in_channels)
        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):
        h_ = x
        h_ = self.norm(h_)
        q = self.q(h_)
        k = self.k(h_)
        v = self.v(h_)

        # compute attention
        b, c, h, w = q.shape
        q = rearrange(q, "b c h w -> b (h w) c")
        k = rearrange(k, "b c h w -> b c (h w)")
        w_ = torch.einsum("bij,bjk->bik", q, k)

        w_ = w_ * (int(c) ** (-0.5))
        w_ = torch.nn.functional.softmax(w_, dim=2)

        # attend to values
        v = rearrange(v, "b c h w -> b c (h w)")
        w_ = rearrange(w_, "b i j -> b j i")
        h_ = torch.einsum("bij,bjk->bik", v, w_)
        h_ = rearrange(h_, "b c (h w) -> b c h w", h=h)
        h_ = self.proj_out(h_)

        return x + h_


def safe_memory_efficient_attention(q, k, v, attn_bias=None, op=None, p=0.0):
    if q.device.type == "cuda" and HAS_XFORMERS:
        return xformers.ops.memory_efficient_attention(q, k, v, attn_bias=attn_bias, op=op, p=p)
    else:
        # Standard attention for CPU
        scale = 1.0 / (q.shape[-1] ** 0.5)
        attn = torch.matmul(q * scale, k.transpose(-2, -1))
        if attn_bias is not None:
            attn = attn + attn_bias
        attn = torch.softmax(attn, dim=-1)
        attn = torch.nn.functional.dropout(attn, p=p)
        return torch.matmul(attn, v)


class MemoryEfficientCrossAttention(nn.Module):
    # https://github.com/MatthieuTPHR/diffusers/blob/d80b531ff8060ec1ea982b65a1b8df70f73aa67c/src/diffusers/models/attention.py#L223
    def __init__(self, query_dim, context_dim=None, heads=8, dim_head=64, dropout=0.0, **kwargs):
        super().__init__()
        print(
            f"Setting up {self.__class__.__name__}. Query dim is {query_dim}, context_dim is {context_dim} and using "
            f"{heads} heads."
        )
        inner_dim = dim_head * heads
        context_dim = default(context_dim, query_dim)

        self.heads = heads
        self.dim_head = dim_head
        
        self.with_ip = kwargs.get("with_ip", False)
        if self.with_ip and (context_dim is not None):
            self.to_k_ip = nn.Linear(context_dim, inner_dim, bias=False)
            self.to_v_ip = nn.Linear(context_dim, inner_dim, bias=False)
            self.ip_dim= kwargs.get("ip_dim", 16)
            self.ip_weight = kwargs.get("ip_weight", 1.0)

        self.to_q = nn.Linear(query_dim, inner_dim, bias=False)
        self.to_k = nn.Linear(context_dim, inner_dim, bias=False)
        self.to_v = nn.Linear(context_dim, inner_dim, bias=False)

        self.to_out = nn.Sequential(
            nn.Linear(inner_dim, query_dim), nn.Dropout(dropout)
        )
        self.attention_op: Optional[Any] = None

    def forward(self, x, context=None, mask=None):
        q = self.to_q(x)
        
        has_ip = self.with_ip and (context is not None)
        if has_ip:
            # context dim [(b frame_num), (77 + img_token), 1024]
            token_len = context.shape[1]
            context_ip = context[:, -self.ip_dim:, :]
            k_ip = self.to_k_ip(context_ip)
            v_ip = self.to_v_ip(context_ip)
            context = context[:, :(token_len - self.ip_dim), :]

        context = default(context, x)
        k = self.to_k(context)
        v = self.to_v(context)

        b, _, _ = q.shape
        q, k, v = map(
            lambda t: t.unsqueeze(3)
            .reshape(b, t.shape[1], self.heads, self.dim_head)
            .permute(0, 2, 1, 3)
            .reshape(b * self.heads, t.shape[1], self.dim_head)
            .contiguous(),
            (q, k, v),
        )

        # actually compute the attention, what we cannot get enough of
        out = safe_memory_efficient_attention(
            q, k, v, attn_bias=None, op=self.attention_op
        )
        
        if has_ip:
            k_ip, v_ip = map(
                lambda t: t.unsqueeze(3)
                .reshape(b, t.shape[1], self.heads, self.dim_head)
                .permute(0, 2, 1, 3)
                .reshape(b * self.heads, t.shape[1], self.dim_head)
                .contiguous(),
                (k_ip, v_ip),
            )
            # actually compute the attention, what we cannot get enough of
            out_ip = safe_memory_efficient_attention(
                q, k_ip, v_ip, attn_bias=None, op=self.attention_op
            )
            out = out + self.ip_weight * out_ip
            
        if exists(mask):
            raise NotImplementedError
        out = (
            out.unsqueeze(0)
            .reshape(b, self.heads, out.shape[1], self.dim_head)
            .permute(0, 2, 1, 3)
            .reshape(b, out.shape[1], self.heads * self.dim_head)
        )
        return self.to_out(out)


class BasicTransformerBlock(nn.Module):
    def __init__(
        self,
        dim,
        n_heads,
        d_head,
        dropout=0.0,
        context_dim=None,
        gated_ff=True,
        checkpoint=True,
        disable_self_attn=False,
        **kwargs
    ):
        super().__init__()
        assert HAS_XFORMERS, "xformers is not available"
        attn_cls = MemoryEfficientCrossAttention
        self.disable_self_attn = disable_self_attn
        self.attn1 = attn_cls(
            query_dim=dim,
            heads=n_heads,
            dim_head=d_head,
            dropout=dropout,
            context_dim=context_dim if self.disable_self_attn else None,
        )  # is a self-attention if not self.disable_self_attn
        self.ff = FeedForward(dim, dropout=dropout, glu=gated_ff)
        self.attn2 = attn_cls(
            query_dim=dim,
            context_dim=context_dim,
            heads=n_heads,
            dim_head=d_head,
            dropout=dropout,
            **kwargs
        )  # is self-attn if context is none
        self.norm1 = nn.LayerNorm(dim)
        self.norm2 = nn.LayerNorm(dim)
        self.norm3 = nn.LayerNorm(dim)
        self.checkpoint = checkpoint

    def forward(self, x, context=None):
        return checkpoint(
            self._forward, (x, context), self.parameters(), self.checkpoint
        )

    def _forward(self, x, context=None):
        x = (
            self.attn1(
                self.norm1(x), context=context if self.disable_self_attn else None
            )
            + x
        )
        x = self.attn2(self.norm2(x), context=context) + x
        x = self.ff(self.norm3(x)) + x
        return x


class SpatialTransformer(nn.Module):
    """
    Transformer block for image-like data.
    First, project the input (aka embedding)
    and reshape to b, t, d.
    Then apply standard transformer action.
    Finally, reshape to image
    NEW: use_linear for more efficiency instead of the 1x1 convs
    """

    def __init__(
        self,
        in_channels,
        n_heads,
        d_head,
        depth=1,
        dropout=0.0,
        context_dim=None,
        disable_self_attn=False,
        use_linear=False,
        use_checkpoint=True,
        **kwargs
    ):
        super().__init__()
        if exists(context_dim) and not isinstance(context_dim, list):
            context_dim = [context_dim]
        self.in_channels = in_channels
        inner_dim = n_heads * d_head
        self.norm = Normalize(in_channels)
        if not use_linear:
            self.proj_in = nn.Conv2d(
                in_channels, inner_dim, kernel_size=1, stride=1, padding=0
            )
        else:
            self.proj_in = nn.Linear(in_channels, inner_dim)

        self.transformer_blocks = nn.ModuleList(
            [
                BasicTransformerBlock(
                    inner_dim,
                    n_heads,
                    d_head,
                    dropout=dropout,
                    context_dim=context_dim[d],
                    disable_self_attn=disable_self_attn,
                    checkpoint=use_checkpoint,
                    **kwargs
                )
                for d in range(depth)
            ]
        )
        if not use_linear:
            self.proj_out = zero_module(
                nn.Conv2d(inner_dim, in_channels, kernel_size=1, stride=1, padding=0)
            )
        else:
            self.proj_out = zero_module(nn.Linear(in_channels, inner_dim))
        self.use_linear = use_linear

    def forward(self, x, context=None):
        # note: if no context is given, cross-attention defaults to self-attention
        if not isinstance(context, list):
            context = [context]
        b, c, h, w = x.shape
        x_in = x
        x = self.norm(x)
        if not self.use_linear:
            x = self.proj_in(x)
        x = rearrange(x, "b c h w -> b (h w) c").contiguous()
        if self.use_linear:
            x = self.proj_in(x)
        for i, block in enumerate(self.transformer_blocks):
            x = block(x, context=context[i])
        if self.use_linear:
            x = self.proj_out(x)
        x = rearrange(x, "b (h w) c -> b c h w", h=h, w=w).contiguous()
        if not self.use_linear:
            x = self.proj_out(x)
        return x + x_in


class BasicTransformerBlock3D(BasicTransformerBlock):
    def forward(self, x, context=None, num_frames=1):
        return checkpoint(
            self._forward, (x, context, num_frames), self.parameters(), self.checkpoint
        )

    def _forward(self, x, context=None, num_frames=1):
        x = rearrange(x, "(b f) l c -> b (f l) c", f=num_frames).contiguous()
        x = (
            self.attn1(
                self.norm1(x), 
                context=context if self.disable_self_attn else None
            )
            + x
        )
        x = rearrange(x, "b (f l) c -> (b f) l c", f=num_frames).contiguous()
        x = self.attn2(self.norm2(x), context=context) + x
        x = self.ff(self.norm3(x)) + x
        return x


class SpatialTransformer3D(nn.Module):
    """3D self-attention"""

    def __init__(
        self,
        in_channels,
        n_heads,
        d_head,
        depth=1,
        dropout=0.0,
        context_dim=None,
        disable_self_attn=False,
        use_linear=False,
        use_checkpoint=True,
        **kwargs
    ):
        super().__init__()
        if exists(context_dim) and not isinstance(context_dim, list):
            context_dim = [context_dim]
        self.in_channels = in_channels
        inner_dim = n_heads * d_head
        self.norm = Normalize(in_channels)
        if not use_linear:
            self.proj_in = nn.Conv2d(
                in_channels, inner_dim, kernel_size=1, stride=1, padding=0
            )
        else:
            self.proj_in = nn.Linear(in_channels, inner_dim)

        self.transformer_blocks = nn.ModuleList(
            [
                BasicTransformerBlock3D(
                    inner_dim,
                    n_heads,
                    d_head,
                    dropout=dropout,
                    context_dim=context_dim[d],
                    disable_self_attn=disable_self_attn,
                    checkpoint=use_checkpoint,
                    **kwargs
                )
                for d in range(depth)
            ]
        )
        if not use_linear:
            self.proj_out = zero_module(
                nn.Conv2d(inner_dim, in_channels, kernel_size=1, stride=1, padding=0)
            )
        else:
            self.proj_out = zero_module(nn.Linear(in_channels, inner_dim))
        self.use_linear = use_linear

    def forward(self, x, context=None, num_frames=1):
        # note: if no context is given, cross-attention defaults to self-attention
        if not isinstance(context, list):
            context = [context]
        b, c, h, w = x.shape
        x_in = x
        x = self.norm(x)
        if not self.use_linear:
            x = self.proj_in(x)
        x = rearrange(x, "b c h w -> b (h w) c").contiguous()
        if self.use_linear:
            x = self.proj_in(x)
        for i, block in enumerate(self.transformer_blocks):
            x = block(x, context=context[i], num_frames=num_frames)
        if self.use_linear:
            x = self.proj_out(x)
        x = rearrange(x, "b (h w) c -> b c h w", h=h, w=w).contiguous()
        if not self.use_linear:
            x = self.proj_out(x)
        return x + x_in