diff --git a/backend/nn/flux.py b/backend/nn/flux.py index 67f888e4..256684cb 100644 --- a/backend/nn/flux.py +++ b/backend/nn/flux.py @@ -1,12 +1,13 @@ -# Single File Implementation of Flux, by Forge +# Single File Implementation of Flux with aggressive optimizations, copyright Forge 2024 +# If used outside Forge, only non-commercial use is allowed. # See also https://github.com/black-forest-labs/flux import math import torch -from einops import rearrange, repeat + from torch import nn -from dataclasses import dataclass +from einops import rearrange, repeat from backend.attention import attention_function @@ -30,6 +31,7 @@ def apply_rope(xq, xk, freqs_cis): xk_ = xk.float().reshape(*xk.shape[:-1], -1, 1, 2) xq_out = freqs_cis[..., 0] * xq_[..., 0] + freqs_cis[..., 1] * xq_[..., 1] xk_out = freqs_cis[..., 0] * xk_[..., 0] + freqs_cis[..., 1] * xk_[..., 1] + del xq_, xk_ return xq_out.reshape(*xq.shape).type_as(xq), xk_out.reshape(*xk.shape).type_as(xk) @@ -38,7 +40,9 @@ def timestep_embedding(t, dim, max_period=10000, time_factor=1000.0): half = dim // 2 freqs = torch.exp(-math.log(max_period) * torch.arange(start=0, end=half, dtype=torch.float32) / half).to(t.device) args = t[:, None].float() * freqs[None] + del freqs embedding = torch.cat([torch.cos(args), torch.sin(args)], dim=-1) + del args if dim % 2: embedding = torch.cat([embedding, torch.zeros_like(embedding[:, :1])], dim=-1) if torch.is_floating_point(t): @@ -46,29 +50,6 @@ def timestep_embedding(t, dim, max_period=10000, time_factor=1000.0): return embedding -@dataclass -class ModulationOut: - shift: torch.Tensor - scale: torch.Tensor - gate: torch.Tensor - - -@dataclass -class FluxParams: - in_channels: int - vec_in_dim: int - context_in_dim: int - hidden_size: int - mlp_ratio: float - num_heads: int - depth: int - depth_single_blocks: int - axes_dim: list[int] - theta: int - qkv_bias: bool - guidance_embed: bool - - class EmbedND(nn.Module): def __init__(self, dim, theta, axes_dim): super().__init__() @@ -82,6 +63,7 @@ class EmbedND(nn.Module): [rope(ids[..., i], self.axes_dim[i], self.theta) for i in range(n_axes)], dim=-3, ) + del ids, n_axes return emb.unsqueeze(1) @@ -93,7 +75,8 @@ class MLPEmbedder(nn.Module): self.out_layer = nn.Linear(hidden_dim, hidden_dim, bias=True) def forward(self, x): - return self.out_layer(self.silu(self.in_layer(x))) + x = self.silu(self.in_layer(x)) + return self.out_layer(x) class RMSNorm(nn.Module): @@ -117,7 +100,8 @@ class QKNorm(nn.Module): def forward(self, q, k, v): q = self.query_norm(q) k = self.key_norm(k) - return q.to(v), k.to(v) + del v + return q.to(k), k.to(q) class SelfAttention(nn.Module): @@ -132,8 +116,10 @@ class SelfAttention(nn.Module): def forward(self, x, pe): qkv = self.qkv(x) q, k, v = rearrange(qkv, "B L (K H D) -> K B H L D", K=3, H=self.num_heads) + del qkv q, k = self.norm(q, k, v) x = attention(q, k, v, pe=pe) + del q, k, v x = self.proj(x) return x @@ -147,10 +133,10 @@ class Modulation(nn.Module): def forward(self, vec): out = self.lin(nn.functional.silu(vec))[:, None, :].chunk(self.multiplier, dim=-1) - return ( - ModulationOut(*out[:3]), - ModulationOut(*out[3:]) if self.is_double else None, - ) + if self.is_double: + return out[0], out[1], out[2], out[3], out[4], out[5], + else: + return out[0], out[1], out[2] class DoubleStreamBlock(nn.Module): @@ -179,27 +165,48 @@ class DoubleStreamBlock(nn.Module): ) def forward(self, img, txt, vec, pe): - img_mod1, img_mod2 = self.img_mod(vec) - txt_mod1, txt_mod2 = self.txt_mod(vec) + img_mod1_shift, img_mod1_scale, img_mod1_gate, img_mod2_shift, img_mod2_scale, img_mod2_gate = self.img_mod(vec) + txt_mod1_shift, txt_mod1_scale, txt_mod1_gate, txt_mod2_shift, txt_mod2_scale, txt_mod2_gate = self.txt_mod(vec) + img_modulated = self.img_norm1(img) - img_modulated = (1 + img_mod1.scale) * img_modulated + img_mod1.shift + img_modulated = (1 + img_mod1_scale) * img_modulated + img_mod1_shift + del img_mod1_shift, img_mod1_scale img_qkv = self.img_attn.qkv(img_modulated) + del img_modulated img_q, img_k, img_v = rearrange(img_qkv, "B L (K H D) -> K B H L D", K=3, H=self.num_heads) + del img_qkv img_q, img_k = self.img_attn.norm(img_q, img_k, img_v) + txt_modulated = self.txt_norm1(txt) - txt_modulated = (1 + txt_mod1.scale) * txt_modulated + txt_mod1.shift + txt_modulated = (1 + txt_mod1_scale) * txt_modulated + txt_mod1_shift + del txt_mod1_shift, txt_mod1_scale txt_qkv = self.txt_attn.qkv(txt_modulated) + del txt_modulated txt_q, txt_k, txt_v = rearrange(txt_qkv, "B L (K H D) -> K B H L D", K=3, H=self.num_heads) + del txt_qkv txt_q, txt_k = self.txt_attn.norm(txt_q, txt_k, txt_v) + q = torch.cat((txt_q, img_q), dim=2) + del txt_q, img_q k = torch.cat((txt_k, img_k), dim=2) + del txt_k, img_k v = torch.cat((txt_v, img_v), dim=2) + del txt_v, img_v + attn = attention(q, k, v, pe=pe) txt_attn, img_attn = attn[:, : txt.shape[1]], attn[:, txt.shape[1]:] - img = img + img_mod1.gate * self.img_attn.proj(img_attn) - img = img + img_mod2.gate * self.img_mlp((1 + img_mod2.scale) * self.img_norm2(img) + img_mod2.shift) - txt = txt + txt_mod1.gate * self.txt_attn.proj(txt_attn) - txt = txt + txt_mod2.gate * self.txt_mlp((1 + txt_mod2.scale) * self.txt_norm2(txt) + txt_mod2.shift) + del attn + + img = img + img_mod1_gate * self.img_attn.proj(img_attn) + del img_attn, img_mod1_gate + img = img + img_mod2_gate * self.img_mlp((1 + img_mod2_scale) * self.img_norm2(img) + img_mod2_shift) + del img_mod2_gate, img_mod2_scale, img_mod2_shift + + txt = txt + txt_mod1_gate * self.txt_attn.proj(txt_attn) + del txt_attn, txt_mod1_gate + txt = txt + txt_mod2_gate * self.txt_mlp((1 + txt_mod2_scale) * self.txt_norm2(txt) + txt_mod2_shift) + del txt_mod2_gate, txt_mod2_scale, txt_mod2_shift + return img, txt @@ -220,14 +227,20 @@ class SingleStreamBlock(nn.Module): self.modulation = Modulation(hidden_size, double=False) def forward(self, x, vec, pe): - mod, _ = self.modulation(vec) - x_mod = (1 + mod.scale) * self.pre_norm(x) + mod.shift + mod_shift, mod_scale, mod_gate = self.modulation(vec) + del vec + x_mod = (1 + mod_scale) * self.pre_norm(x) + mod_shift + del mod_shift, mod_scale qkv, mlp = torch.split(self.linear1(x_mod), [3 * self.hidden_size, self.mlp_hidden_dim], dim=-1) + del x_mod q, k, v = rearrange(qkv, "B L (K H D) -> K B H L D", K=3, H=self.num_heads) + del qkv q, k = self.norm(q, k, v) attn = attention(q, k, v, pe=pe) + del q, k, v, pe output = self.linear2(torch.cat((attn, self.mlp_act(mlp)), 2)) - return x + mod.gate * output + del attn, mlp + return x + mod_gate * output class LastLayer(nn.Module): @@ -239,48 +252,56 @@ class LastLayer(nn.Module): def forward(self, x, vec): shift, scale = self.adaLN_modulation(vec).chunk(2, dim=1) + del vec x = (1 + scale[:, None, :]) * self.norm_final(x) + shift[:, None, :] + del scale, shift x = self.linear(x) return x class IntegratedFluxTransformer2DModel(nn.Module): - def __init__(self, **kwargs): + def __init__(self, in_channels: int, vec_in_dim: int, context_in_dim: int, hidden_size: int, mlp_ratio: float, num_heads: int, depth: int, depth_single_blocks: int, axes_dim: list[int], theta: int, qkv_bias: bool, guidance_embed: bool): super().__init__() - params = FluxParams(**kwargs) - self.params = params - self.in_channels = params.in_channels * 4 + + self.in_channels = in_channels * 4 self.out_channels = self.in_channels - if params.hidden_size % params.num_heads != 0: - raise ValueError(f"Hidden size {params.hidden_size} must be divisible by num_heads {params.num_heads}") - pe_dim = params.hidden_size // params.num_heads - if sum(params.axes_dim) != pe_dim: - raise ValueError(f"Got {params.axes_dim} but expected positional dim {pe_dim}") - self.hidden_size = params.hidden_size - self.num_heads = params.num_heads - self.pe_embedder = EmbedND(dim=pe_dim, theta=params.theta, axes_dim=params.axes_dim) + + if hidden_size % num_heads != 0: + raise ValueError(f"Hidden size {hidden_size} must be divisible by num_heads {num_heads}") + + pe_dim = hidden_size // num_heads + if sum(axes_dim) != pe_dim: + raise ValueError(f"Got {axes_dim} but expected positional dim {pe_dim}") + + self.hidden_size = hidden_size + self.num_heads = num_heads + + self.pe_embedder = EmbedND(dim=pe_dim, theta=theta, axes_dim=axes_dim) self.img_in = nn.Linear(self.in_channels, self.hidden_size, bias=True) self.time_in = MLPEmbedder(in_dim=256, hidden_dim=self.hidden_size) - self.vector_in = MLPEmbedder(params.vec_in_dim, self.hidden_size) - self.guidance_in = MLPEmbedder(in_dim=256, hidden_dim=self.hidden_size) if params.guidance_embed else nn.Identity() - self.txt_in = nn.Linear(params.context_in_dim, self.hidden_size) + self.vector_in = MLPEmbedder(vec_in_dim, self.hidden_size) + self.guidance_in = MLPEmbedder(in_dim=256, hidden_dim=self.hidden_size) if guidance_embed else nn.Identity() + self.txt_in = nn.Linear(context_in_dim, self.hidden_size) + self.double_blocks = nn.ModuleList( [ DoubleStreamBlock( self.hidden_size, self.num_heads, - mlp_ratio=params.mlp_ratio, - qkv_bias=params.qkv_bias, + mlp_ratio=mlp_ratio, + qkv_bias=qkv_bias, ) - for _ in range(params.depth) + for _ in range(depth) ] ) + self.single_blocks = nn.ModuleList( [ - SingleStreamBlock(self.hidden_size, self.num_heads, mlp_ratio=params.mlp_ratio) - for _ in range(params.depth_single_blocks) + SingleStreamBlock(self.hidden_size, self.num_heads, mlp_ratio=mlp_ratio) + for _ in range(depth_single_blocks) ] ) + self.final_layer = LastLayer(self.hidden_size, 1, self.out_channels) def inner_forward(self, img, img_ids, txt, txt_ids, timesteps, y, guidance=None): @@ -294,15 +315,21 @@ class IntegratedFluxTransformer2DModel(nn.Module): vec = vec + self.guidance_in(timestep_embedding(guidance, 256).to(img.dtype)) vec = vec + self.vector_in(y) txt = self.txt_in(txt) + del y, guidance ids = torch.cat((txt_ids, img_ids), dim=1) + del txt_ids, img_ids pe = self.pe_embedder(ids) + del ids for block in self.double_blocks: img, txt = block(img=img, txt=txt, vec=vec, pe=pe) img = torch.cat((txt, img), 1) for block in self.single_blocks: img = block(img, vec=vec, pe=pe) + del pe img = img[:, txt.shape[1]:, ...] + del txt img = self.final_layer(img, vec) + del vec return img def forward(self, x, timestep, context, y, guidance, **kwargs): @@ -314,6 +341,7 @@ class IntegratedFluxTransformer2DModel(nn.Module): pad_w = (patch_size - x.shape[-1] % patch_size) % patch_size x = torch.nn.functional.pad(x, (0, pad_w, 0, pad_h), mode="circular") img = rearrange(x, "b c (h ph) (w pw) -> b (h w) (c ph pw)", ph=patch_size, pw=patch_size) + del x, pad_h, pad_w h_len = ((h + (patch_size // 2)) // patch_size) w_len = ((w + (patch_size // 2)) // patch_size) img_ids = torch.zeros((h_len, w_len, 3), device=input_device, dtype=input_dtype) @@ -321,6 +349,9 @@ class IntegratedFluxTransformer2DModel(nn.Module): img_ids[..., 2] = img_ids[..., 2] + torch.linspace(0, w_len - 1, steps=w_len, device=input_device, dtype=input_dtype)[None, :] img_ids = repeat(img_ids, "h w c -> b (h w) c", b=bs) txt_ids = torch.zeros((bs, context.shape[1], 3), device=input_device, dtype=input_dtype) + del input_device, input_dtype out = self.inner_forward(img, img_ids, context, txt_ids, timestep, y, guidance) + del img, img_ids, txt_ids, timestep, context out = rearrange(out, "b (h w) (c ph pw) -> b c (h ph) (w pw)", h=h_len, w=w_len, ph=2, pw=2)[:, :, :h, :w] + del h_len, w_len, bs return out