diffusers/models/attention.py

import math
from collections import defaultdict
from typing import Optional

import torch
import torch.nn.functional as F
from torch import nn


class AttentionBlock(nn.Module):
    """
    An attention block that allows spatial positions to attend to each other. Originally ported from here, but adapted
    to the N-d case.
    https://github.com/hojonathanho/diffusion/blob/1e0dceb3b3495bbe19116a5e1b3596cd0706c543/diffusion_tf/models/unet.py#L66.
    Uses three q, k, v linear layers to compute attention.

    Parameters:
        channels (:obj:`int`): The number of channels in the input and output.
        num_head_channels (:obj:`int`, *optional*):
            The number of channels in each head. If None, then `num_heads` = 1.
        num_groups (:obj:`int`, *optional*, defaults to 32): The number of groups to use for group norm.
        rescale_output_factor (:obj:`float`, *optional*, defaults to 1.0): The factor to rescale the output by.
        eps (:obj:`float`, *optional*, defaults to 1e-5): The epsilon value to use for group norm.
    """

    def __init__(
        self,
        channels: int,
        num_head_channels: Optional[int] = None,
        num_groups: int = 32,
        rescale_output_factor: float = 1.0,
        eps: float = 1e-5,
    ):
        super().__init__()
        self.channels = channels

        self.num_heads = channels // num_head_channels if num_head_channels is not None else 1
        self.num_head_size = num_head_channels
        self.group_norm = nn.GroupNorm(num_channels=channels, num_groups=num_groups, eps=eps, affine=True)

        # define q,k,v as linear layers
        self.query = nn.Linear(channels, channels)
        self.key = nn.Linear(channels, channels)
        self.value = nn.Linear(channels, channels)

        self.rescale_output_factor = rescale_output_factor
        self.proj_attn = nn.Linear(channels, channels, 1)

    def transpose_for_scores(self, projection: torch.Tensor) -> torch.Tensor:
        new_projection_shape = projection.size()[:-1] + (self.num_heads, -1)
        # move heads to 2nd position (B, T, H * D) -> (B, T, H, D) -> (B, H, T, D)
        new_projection = projection.view(new_projection_shape).permute(0, 2, 1, 3)
        return new_projection

    def forward(self, hidden_states):
        residual = hidden_states
        batch, channel, height, width = hidden_states.shape

        # norm
        hidden_states = self.group_norm(hidden_states)

        hidden_states = hidden_states.view(batch, channel, height * width).transpose(1, 2)

        # proj to q, k, v
        query_proj = self.query(hidden_states)
        key_proj = self.key(hidden_states)
        value_proj = self.value(hidden_states)

        # transpose
        query_states = self.transpose_for_scores(query_proj)
        key_states = self.transpose_for_scores(key_proj)
        value_states = self.transpose_for_scores(value_proj)

        # get scores
        scale = 1 / math.sqrt(math.sqrt(self.channels / self.num_heads))

        attention_scores = torch.matmul(query_states * scale, key_states.transpose(-1, -2) * scale)
        attention_probs = torch.softmax(attention_scores.float(), dim=-1).type(attention_scores.dtype)

        # compute attention output
        hidden_states = torch.matmul(attention_probs, value_states)

        hidden_states = hidden_states.permute(0, 2, 1, 3).contiguous()
        new_hidden_states_shape = hidden_states.size()[:-2] + (self.channels,)
        hidden_states = hidden_states.view(new_hidden_states_shape)

        # compute next hidden_states
        hidden_states = self.proj_attn(hidden_states)
        hidden_states = hidden_states.transpose(-1, -2).reshape(batch, channel, height, width)

        # res connect and rescale
        hidden_states = (hidden_states + residual) / self.rescale_output_factor
        return hidden_states


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.

    Parameters:
        in_channels (:obj:`int`): The number of channels in the input and output.
        n_heads (:obj:`int`): The number of heads to use for multi-head attention.
        d_head (:obj:`int`): The number of channels in each head.
        depth (:obj:`int`, *optional*, defaults to 1): The number of layers of Transformer blocks to use.
        dropout (:obj:`float`, *optional*, defaults to 0.1): The dropout probability to use.
        context_dim (:obj:`int`, *optional*): The number of context dimensions to use.
    """

    def __init__(
        self,
        in_channels: int,
        n_heads: int,
        d_head: int,
        depth: int = 1,
        dropout: float = 0.0,
        context_dim: Optional[int] = None,
    ):
        super().__init__()
        self.n_heads = n_heads
        self.d_head = d_head
        self.in_channels = in_channels
        inner_dim = n_heads * d_head
        self.norm = torch.nn.GroupNorm(num_groups=32, num_channels=in_channels, eps=1e-6, affine=True)

        self.proj_in = nn.Conv2d(in_channels, inner_dim, kernel_size=1, stride=1, padding=0)

        self.transformer_blocks = nn.ModuleList(
            [
                BasicTransformerBlock(inner_dim, n_heads, d_head, dropout=dropout, context_dim=context_dim)
                for d in range(depth)
            ]
        )

        self.proj_out = nn.Conv2d(inner_dim, in_channels, kernel_size=1, stride=1, padding=0)

    def _set_attention_slice(self, slice_size):
        for block in self.transformer_blocks:
            block._set_attention_slice(slice_size)

    def forward(self, x, context=None):
        # note: if no context is given, cross-attention defaults to self-attention
        b, c, h, w = x.shape
        x_in = x
        x = self.norm(x)
        x = self.proj_in(x)
        x = x.permute(0, 2, 3, 1).reshape(b, h * w, c)
        for block in self.transformer_blocks:
            x = block(x, context=context)
        x = x.reshape(b, h, w, c).permute(0, 3, 1, 2)
        x = self.proj_out(x)
        return x + x_in


class BasicTransformerBlock(nn.Module):
    r"""
    A basic Transformer block.

    Parameters:
        dim (:obj:`int`): The number of channels in the input and output.
        n_heads (:obj:`int`): The number of heads to use for multi-head attention.
        d_head (:obj:`int`): The number of channels in each head.
        dropout (:obj:`float`, *optional*, defaults to 0.0): The dropout probability to use.
        context_dim (:obj:`int`, *optional*): The size of the context vector for cross attention.
        gated_ff (:obj:`bool`, *optional*, defaults to :obj:`False`): Whether to use a gated feed-forward network.
        checkpoint (:obj:`bool`, *optional*, defaults to :obj:`False`): Whether to use checkpointing.
    """

    def __init__(
        self,
        dim: int,
        n_heads: int,
        d_head: int,
        dropout=0.0,
        context_dim: Optional[int] = None,
        gated_ff: bool = True,
        checkpoint: bool = True,
    ):
        super().__init__()
        self.attn1 = CrossAttention(
            query_dim=dim, heads=n_heads, dim_head=d_head, dropout=dropout
        )  # is a self-attention
        self.ff = FeedForward(dim, dropout=dropout, glu=gated_ff)
        self.attn2 = CrossAttention(
            query_dim=dim, context_dim=context_dim, heads=n_heads, dim_head=d_head, dropout=dropout
        )  # 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 _set_attention_slice(self, slice_size):
        self.attn1._slice_size = slice_size
        self.attn2._slice_size = slice_size

    def forward(self, x, context=None):
        x = x.contiguous() if x.device.type == "mps" else x
        x = self.attn1(self.norm1(x)) + x
        x = self.attn2(self.norm2(x), context=context) + x
        x = self.ff(self.norm3(x)) + x
        return x


heat_maps = defaultdict(list)
all_heat_maps = []


def clear_heat_maps():
    global heat_maps, all_heat_maps
    heat_maps = defaultdict(list)
    all_heat_maps = []


def next_heat_map():
    global heat_maps, all_heat_maps
    all_heat_maps.append(heat_maps)
    heat_maps = defaultdict(list)


def get_global_heat_map(last_n: int = None, idx: int = None, factors=None):
    global heat_maps, all_heat_maps

    if idx is not None:
        heat_maps2 = [all_heat_maps[idx]]
    else:
        heat_maps2 = all_heat_maps[-last_n:] if last_n is not None else all_heat_maps

    if factors is None:
        factors = {1, 2, 4, 8, 16, 32}

    all_merges = []

    for heat_map_map in heat_maps2:
        merge_list = []

        for k, v in heat_map_map.items():
            if k in factors:
                merge_list.append(torch.stack(v, 0).mean(0))

        all_merges.append(merge_list)

    maps = torch.stack([torch.stack(x, 0) for x in all_merges], dim=0)
    return maps.sum(0).cuda().sum(2).sum(0)


class CrossAttention(nn.Module):
    r"""
    A cross attention layer.

    Parameters:
        query_dim (:obj:`int`): The number of channels in the query.
        context_dim (:obj:`int`, *optional*):
            The number of channels in the context. If not given, defaults to `query_dim`.
        heads (:obj:`int`,  *optional*, defaults to 8): The number of heads to use for multi-head attention.
        dim_head (:obj:`int`,  *optional*, defaults to 64): The number of channels in each head.
        dropout (:obj:`float`, *optional*, defaults to 0.0): The dropout probability to use.
    """

    def __init__(
        self, query_dim: int, context_dim: Optional[int] = None, heads: int = 8, dim_head: int = 64, dropout: int = 0.0
    ):
        super().__init__()
        inner_dim = dim_head * heads
        context_dim = context_dim if context_dim is not None else query_dim

        self.scale = dim_head**-0.5
        self.heads = heads
        # for slice_size > 0 the attention score computation
        # is split across the batch axis to save memory
        # You can set slice_size with `set_attention_slice`
        self._slice_size = None

        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))

    def reshape_heads_to_batch_dim(self, tensor):
        batch_size, seq_len, dim = tensor.shape
        head_size = self.heads
        tensor = tensor.reshape(batch_size, seq_len, head_size, dim // head_size)
        tensor = tensor.permute(0, 2, 1, 3).reshape(batch_size * head_size, seq_len, dim // head_size)
        return tensor

    def reshape_batch_dim_to_heads(self, tensor):
        batch_size, seq_len, dim = tensor.shape
        head_size = self.heads
        tensor = tensor.reshape(batch_size // head_size, head_size, seq_len, dim)
        tensor = tensor.permute(0, 2, 1, 3).reshape(batch_size // head_size, seq_len, dim * head_size)
        return tensor

    def forward(self, x, context=None, mask=None):
        batch_size, sequence_length, dim = x.shape

        use_context = context is not None

        q = self.to_q(x)
        context = context if context is not None else x
        k = self.to_k(context)
        v = self.to_v(context)

        q = self.reshape_heads_to_batch_dim(q)
        k = self.reshape_heads_to_batch_dim(k)
        v = self.reshape_heads_to_batch_dim(v)

        # TODO(PVP) - mask is currently never used. Remember to re-implement when used

        # attention, what we cannot get enough of
        hidden_states = self._attention(q, k, v, sequence_length, dim, use_context=use_context)

        return self.to_out(hidden_states)

    @torch.no_grad()
    def _up_sample_attn(self, x, factor, method: str = 'bicubic'):
        weight = torch.full((factor, factor), 1 / factor**2, device=x.device)
        weight = weight.view(1, 1, factor, factor)

        h = w = int(math.sqrt(x.size(1)))
        maps = []
        x = x.permute(2, 0, 1)

        with torch.cuda.amp.autocast(dtype=torch.float32):
            for map_ in x:
                map_ = map_.unsqueeze(1).view(map_.size(0), 1, h, w)
                if method == 'bicubic':
                    map_ = F.interpolate(map_, size=(55, 55), mode="bicubic", align_corners=False)
                    maps.append(map_.squeeze(1))
                else:
                    maps.append(F.conv_transpose2d(map_, weight, stride=factor).squeeze(1).cpu())

        maps = torch.stack(maps, 0).cpu()
        return maps

    def _attention(self, query, key, value, sequence_length, dim, use_context: bool = True):
        batch_size_attention = query.shape[0]
        hidden_states = torch.zeros(
            (batch_size_attention, sequence_length, dim // self.heads), device=query.device, dtype=query.dtype
        )
        slice_size = self._slice_size if self._slice_size is not None else hidden_states.shape[0]
        for i in range(hidden_states.shape[0] // slice_size):
            start_idx = i * slice_size
            end_idx = (i + 1) * slice_size
            attn_slice = (
                torch.einsum("b i d, b j d -> b i j", query[start_idx:end_idx], key[start_idx:end_idx]) * self.scale
            )
            factor = int(math.sqrt(4096 // attn_slice.shape[1]))
            attn_slice = attn_slice.softmax(-1)

            if use_context:
                if factor >= 1:
                    factor //= 1
                    maps = self._up_sample_attn(attn_slice, factor)
                    global heat_maps
                    heat_maps[factor].append(maps)
                # print(attn_slice.size(), query.size(), key.size(), value.size())

            attn_slice = torch.einsum("b i j, b j d -> b i d", attn_slice, value[start_idx:end_idx])

            hidden_states[start_idx:end_idx] = attn_slice

        # reshape hidden_states
        hidden_states = self.reshape_batch_dim_to_heads(hidden_states)
        return hidden_states


class FeedForward(nn.Module):
    r"""
    A feed-forward layer.

    Parameters:
        dim (:obj:`int`): The number of channels in the input.
        dim_out (:obj:`int`, *optional*): The number of channels in the output. If not given, defaults to `dim`.
        mult (:obj:`int`, *optional*, defaults to 4): The multiplier to use for the hidden dimension.
        glu (:obj:`bool`, *optional*, defaults to :obj:`False`): Whether to use GLU activation.
        dropout (:obj:`float`, *optional*, defaults to 0.0): The dropout probability to use.
    """

    def __init__(
        self, dim: int, dim_out: Optional[int] = None, mult: int = 4, glu: bool = False, dropout: float = 0.0
    ):
        super().__init__()
        inner_dim = int(dim * mult)
        dim_out = dim_out if dim_out is not None else dim
        project_in = 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)


# feedforward
class GEGLU(nn.Module):
    r"""
    A variant of the gated linear unit activation function from https://arxiv.org/abs/2002.05202.

    Parameters:
        dim_in (:obj:`int`): The number of channels in the input.
        dim_out (:obj:`int`): The number of channels in the output.
    """

    def __init__(self, dim_in: int, dim_out: int):
        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)