openlrm/models/block.py

# Copyright (c) 2023-2024, Zexin He
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.


import torch.nn as nn
import loratorch as lora
from .modulate import ModLN


class BasicBlock(nn.Module):
    """
    Transformer block that is in its simplest form.
    Designed for PF-LRM architecture.
    """
    # Block contains a self-attention layer and an MLP
    def __init__(self, inner_dim: int, num_heads: int, eps: float,
                 attn_drop: float = 0., attn_bias: bool = False,
                 mlp_ratio: float = 4., mlp_drop: float = 0.):
        super().__init__()
        self.norm1 = nn.LayerNorm(inner_dim, eps=eps)
        self.self_attn = nn.MultiheadAttention(
            embed_dim=inner_dim, num_heads=num_heads,
            dropout=attn_drop, bias=attn_bias, batch_first=True)
        self.norm2 = nn.LayerNorm(inner_dim, eps=eps)
        self.mlp = nn.Sequential(
            nn.Linear(inner_dim, int(inner_dim * mlp_ratio)),
            nn.GELU(),
            nn.Dropout(mlp_drop),
            nn.Linear(int(inner_dim * mlp_ratio), inner_dim),
            nn.Dropout(mlp_drop),
        )

    def forward(self, x):
        # x: [N, L, D]
        before_sa = self.norm1(x)
        x = x + self.self_attn(before_sa, before_sa, before_sa, need_weights=False)[0]
        x = x + self.mlp(self.norm2(x))
        return x


class ConditionBlock(nn.Module):
    """
    Transformer block that takes in a cross-attention condition.
    Designed for SparseLRM architecture.
    """
    # Block contains a cross-attention layer, a self-attention layer, and an MLP
    def __init__(self, inner_dim: int, cond_dim: int, num_heads: int, eps: float,
                 attn_drop: float = 0., attn_bias: bool = False,
                 mlp_ratio: float = 4., mlp_drop: float = 0.,
                 lora_rank: int = 0):
        super().__init__()
        self.norm1 = nn.LayerNorm(inner_dim, eps=eps)
        self.cross_attn = lora.MultiheadAttention(
            embed_dim=inner_dim, num_heads=num_heads, kdim=cond_dim, vdim=cond_dim,
            dropout=attn_drop, bias=attn_bias, batch_first=True, r=lora_rank)
        self.norm2 = nn.LayerNorm(inner_dim, eps=eps)
        self.self_attn = lora.MultiheadAttention(
            embed_dim=inner_dim, num_heads=num_heads,
            dropout=attn_drop, bias=attn_bias, batch_first=True, r=lora_rank)
        self.norm3 = nn.LayerNorm(inner_dim, eps=eps)
        self.mlp = nn.Sequential(
            nn.Linear(inner_dim, int(inner_dim * mlp_ratio)),
            nn.GELU(),
            nn.Dropout(mlp_drop),
            nn.Linear(int(inner_dim * mlp_ratio), inner_dim),
            nn.Dropout(mlp_drop),
        )

    def forward(self, x, cond):
        # x: [N, L, D]
        # cond: [N, L_cond, D_cond]
        x = x + self.cross_attn(self.norm1(x), cond, cond, need_weights=False)[0]
        before_sa = self.norm2(x)
        x = x + self.self_attn(before_sa, before_sa, before_sa, need_weights=False)[0]
        x = x + self.mlp(self.norm3(x))
        return x


class ConditionModulationBlock(nn.Module):
    """
    Transformer block that takes in a cross-attention condition and another modulation vector applied to sub-blocks.
    Designed for raw LRM architecture.
    """
    # Block contains a cross-attention layer, a self-attention layer, and an MLP
    def __init__(self, inner_dim: int, cond_dim: int, mod_dim: int, num_heads: int, eps: float,
                 attn_drop: float = 0., attn_bias: bool = False,
                 mlp_ratio: float = 4., mlp_drop: float = 0.,
                 lora_rank: int = 0):
        super().__init__()
        self.norm1 = ModLN(inner_dim, mod_dim, eps)
        self.cross_attn = lora.MultiheadAttention(
            embed_dim=inner_dim, num_heads=num_heads, kdim=cond_dim, vdim=cond_dim,
            dropout=attn_drop, bias=attn_bias, batch_first=True, r=lora_rank)
        self.norm2 = ModLN(inner_dim, mod_dim, eps)
        self.self_attn = lora.MultiheadAttention(
            embed_dim=inner_dim, num_heads=num_heads,
            dropout=attn_drop, bias=attn_bias, batch_first=True, r=lora_rank)
        self.norm3 = ModLN(inner_dim, mod_dim, eps)
        self.mlp = nn.Sequential(
            nn.Linear(inner_dim, int(inner_dim * mlp_ratio)),
            nn.GELU(),
            nn.Dropout(mlp_drop),
            nn.Linear(int(inner_dim * mlp_ratio), inner_dim),
            nn.Dropout(mlp_drop),
        )

    def forward(self, x, cond, mod):
        # x: [N, L, D]
        # cond: [N, L_cond, D_cond]
        # mod: [N, D_mod]
        x = x + self.cross_attn(self.norm1(x, mod), cond, cond, need_weights=False)[0]
        before_sa = self.norm2(x, mod)
        x = x + self.self_attn(before_sa, before_sa, before_sa, need_weights=False)[0]
        x = x + self.mlp(self.norm3(x, mod))
        return x