Article directory
- Self Attention
- Multi-Head Attention
- Cross_MultiAttention
- Cross Attention
- main
Multimodal Conditional Mechanism Cross Attention Principle:
- https://mp.weixin.qq.com/s?__biz=Mzk0MzIzODM5MA== & amp;mid=2247486441 & amp;idx=1 & amp;sn=06df067828b19ef9aeef99f455f897e9 & amp;chksm=c337b670f4403f663f7b98a2aa75cb5062bf5a6222c81ce8f181d79d367971a4587b62da84a1#rd
Self Attention
class SelfAttention(nn.Module):
def __init__(self, emb_dim):
super(SelfAttention, self).__init__()
self.emb_dim = emb_dim
self.Wq = nn.Linear(emb_dim, emb_dim, bias=False)
self.Wk = nn.Linear(emb_dim, emb_dim, bias=False)
self.Wv = nn.Linear(emb_dim, emb_dim, bias=False)
self.fc = nn.Linear(emb_dim, emb_dim)
def forward(self, x, pad_mask=None):
# [batch_szie, seq_len, emb_dim] = [3, 5, 512]
Q = self. Wq(x)
K = self.Wk(x)
V = self.Wv(x)
att_weights = torch.bmm(Q, K.transpose(1, 2)) # [batch_szie, seq_len, seq_len] = [3, 5, 5]
att_weights = att_weights / math.sqrt(self.emb_dim)
if pad_mask is not None:
att_weights = att_weights. masked_fill(pad_mask, -1e9)
att_weights = F.softmax(att_weights, dim=-1)
output = torch.bmm(att_weights, V) # [batch_szie, seq_len, emb_dim] = [3, 5, 512]
output = self.fc(output)
return output, att_weights
Multi-Head Attention
class MultiHeadAttention(nn.Module):
def __init__(self, emb_dim, num_heads, att_dropout=0.0):
super(MultiHeadAttention, self).__init__()
self.emb_dim = emb_dim
self.num_heads = num_heads
self.att_dropout = att_dropout
assert emb_dim % num_heads == 0, "emb_dim must be divisible by num_heads"
self.depth = emb_dim // num_heads
self.Wq = nn.Linear(emb_dim, emb_dim, bias=False)
self.Wk = nn.Linear(emb_dim, emb_dim, bias=False)
self.Wv = nn.Linear(emb_dim, emb_dim, bias=False)
self.fc = nn.Linear(emb_dim, emb_dim)
def forward(self, x, pad_mask=None):
# [batch_szie, seq_len, emb_dim] = [3, 5, 512]
batch_size = x. size(0)
# [batch_szie, seq_len, emb_dim] = [3, 5, 512]
Q = self. Wq(x)
K = self.Wk(x)
V = self.Wv(x)
# Split heads [batch_szie, num_heads, seq_len, depth] = [3, 8, 5, 512/8=64]
Q = Q.view(batch_size, -1, self.num_heads, self.depth).transpose(1, 2)
K = K.view(batch_size, -1, self.num_heads, self.depth).transpose(1, 2)
V = V.view(batch_size, -1, self.num_heads, self.depth).transpose(1, 2)
# [batch_szie, num_heads, seq_len, seq_len] = [3, 8, 5, 5]
att_weights = torch.matmul(Q, K.transpose(-2, -1))
att_weights = att_weights / math. sqrt(self. depth)
if pad_mask is not None:
# Because it is multi-headed, the dimension of the mask matrix should be expanded to 4 dimensions [batch_size, seq_len, seq_len] -> [batch_size, nums_head, seq_len, seq_len]
pad_mask = pad_mask. unsqueeze(1). repeat(1, self. num_heads, 1, 1)
att_weights = att_weights. masked_fill(pad_mask, -1e9)
att_weights = F.softmax(att_weights, dim=-1)
# My own multi-head attention effect is not as good as torch's, I guess because its dropout gives att weight instead of fc
if self.att_dropout > 0.0:
att_weights = F.dropout(att_weights, p=self.att_dropout)
# [batch_szie, num_heads, seq_len, depth] = [3, 8, 5, 64]
output = torch.matmul(att_weights, V)
# Concatenation of results from different heads [batch_szie, seq_len, emb_dim] = [3, 5, 512]
output = output.transpose(1, 2).contiguous().view(batch_size, -1, self.emb_dim)
output = self.fc(output)
return output, att_weights
Cross_MultiAttention
class Cross_MultiAttention(nn.Module):
def __init__(self, in_channels, emb_dim, num_heads, att_dropout=0.0, aropout=0.0):
super(Cross_MultiAttention, self).__init__()
self.emb_dim = emb_dim
self.num_heads = num_heads
self.scale = emb_dim ** -0.5
assert emb_dim % num_heads == 0, "emb_dim must be divisible by num_heads"
self.depth = emb_dim // num_heads
self.proj_in = nn.Conv2d(in_channels, emb_dim, kernel_size=1, stride=1, padding=0)
self.Wq = nn.Linear(emb_dim, emb_dim)
self.Wk = nn.Linear(emb_dim, emb_dim)
self.Wv = nn.Linear(emb_dim, emb_dim)
self.proj_out = nn.Conv2d(emb_dim, in_channels, kernel_size=1, stride=1, padding=0)
def forward(self, x, context, pad_mask=None):
'''
:param x: [batch_size, c, h, w]
:param context: [batch_szie, seq_len, emb_dim]
:param pad_mask: [batch_size, seq_len, seq_len]
:return:
'''
b, c, h, w = x.shape
x = self.proj_in(x) # [batch_size, c, h, w] = [3, 512, 512, 512]
x = rearrange(x, 'b c h w -> b (h w) c') # [batch_size, h*w, c] = [3, 262144, 512]
Q = self.Wq(x) # [batch_size, h*w, emb_dim] = [3, 262144, 512]
K = self.Wk(context) # [batch_szie, seq_len, emb_dim] = [3, 5, 512]
V = self. Wv(context)
Q = Q.view(batch_size, -1, self.num_heads, self.depth).transpose(1, 2) # [batch_size, num_heads, h*w, depth]
K = K.view(batch_size, -1, self.num_heads, self.depth).transpose(1, 2) # [batch_size, num_heads, seq_len, depth]
V = V.view(batch_size, -1, self.num_heads, self.depth).transpose(1, 2)
# [batch_size, num_heads, h*w, seq_len]
att_weights = torch.einsum('bnid,bnjd -> bnij', Q, K)
att_weights = att_weights * self.scale
if pad_mask is not None:
# Because it is multi-headed, the dimension of the mask matrix should be expanded to 4 dimensions [batch_size, h*w, seq_len] -> [batch_size, nums_head, h*w, seq_len]
pad_mask = pad_mask. unsqueeze(1). repeat(1, self. num_heads, 1, 1)
att_weights = att_weights. masked_fill(pad_mask, -1e9)
att_weights = F.softmax(att_weights, dim=-1)
out = torch.einsum('bnij, bnjd -> bnid', att_weights, V)
out = out.transpose(1, 2).contiguous().view(batch_size, -1, self.emb_dim) # [batch_size, h*w, emb_dim]
print(out. shape)
out = rearrange(out, 'b (h w) c -> b c h w', h=h, w=w) # [batch_size, c, h, w]
out = self.proj_out(out) # [batch_size, c, h, w]
return out, att_weights
Cross Attention
class CrossAttention(nn.Module):
def __init__(self, in_channels, emb_dim, att_dropout=0.0, aropout=0.0):
super(CrossAttention, self).__init__()
self.emb_dim = emb_dim
self.scale = emb_dim ** -0.5
self.proj_in = nn.Conv2d(in_channels, emb_dim, kernel_size=1, stride=1, padding=0)
self.Wq = nn.Linear(emb_dim, emb_dim)
self.Wk = nn.Linear(emb_dim, emb_dim)
self.Wv = nn.Linear(emb_dim, emb_dim)
self.proj_out = nn.Conv2d(emb_dim, in_channels, kernel_size=1, stride=1, padding=0)
def forward(self, x, context, pad_mask=None):
'''
:param x: [batch_size, c, h, w]
:param context: [batch_szie, seq_len, emb_dim]
:param pad_mask: [batch_size, seq_len, seq_len]
:return:
'''
b, c, h, w = x.shape
x = self.proj_in(x) # [batch_size, c, h, w] = [3, 512, 512, 512]
x = rearrange(x, 'b c h w -> b (h w) c') # [batch_size, h*w, c] = [3, 262144, 512]
Q = self.Wq(x) # [batch_size, h*w, emb_dim] = [3, 262144, 512]
K = self.Wk(context) # [batch_szie, seq_len, emb_dim] = [3, 5, 512]
V = self. Wv(context)
# [batch_size, h*w, seq_len]
att_weights = torch.einsum('bid,bjd -> bij', Q, K)
att_weights = att_weights * self. scale
if pad_mask is not None:
# [batch_size, h*w, seq_len]
att_weights = att_weights. masked_fill(pad_mask, -1e9)
att_weights = F.softmax(att_weights, dim=-1)
out = torch.einsum('bij, bjd -> bid', att_weights, V) # [batch_size, h*w, emb_dim]
out = rearrange(out, 'b (h w) c -> b c h w', h=h, w=w) # [batch_size, c, h, w]
out = self.proj_out(out) # [batch_size, c, h, w]
print(out. shape)
return out, att_weights
main
# coding:utf-8 # @Email: [email protected] # @Time: 2023/3/22 22:58 # @File: att_test.py ''' Self Attention Multi-Head Attention Cross Attention ''' import torch import torch.nn as nn import torch.nn.functional as F import math from einops import rearrange, repeat from torch.nn import MultiheadAttention if __name__ == '__main__': ''' ''' ''' Assuming that the vocabulary is mapped and entered batch_size = 3 seq_len = max_len = 5 pad = 0 emb_dim = 512 ''' batch_size = 3 seq_len = 5 emb_dim = 512 # In this example, the vocabulary size is 301 vocab_size = 301 input_ids = torch.tensor([[100, 200, 300, 300, 0], [22, 33, 44, 0, 0], [66, 55, 66, 30, 0]], dtype=torch.long) pad_mask = input_ids.eq(0) # logic matrix pad_mask: mark the filling position as True, and mark other positions as False # pad_mask = pad_mask.unsqueeze(1).expand(batch_size, seq_len, seq_len) # [batch_size, seq_len, seq_len] = [3, 5, 5] inputs = nn.Embedding(vocab_size, embedding_dim=emb_dim)(input_ids) # [batch_szie, seq_len, emb_dim] = [3, 5, 512] # self_att = SelfAttention(emb_dim=emb_dim) # self_att(inputs, pad_mask=pad_mask) # multi_att = MultiHeadAttention(emb_dim=emb_dim, num_heads=8) # multi_att(inputs, pad_mask=pad_mask) # Define image data [batch_size, c, h, w] input_img = torch.randn((3, 3, 512, 512)) pad_mask = pad_mask.unsqueeze(1).expand(batch_size, 512*512, seq_len) # cross_att = Cross_MultiAttention(in_channels=3, emb_dim=emb_dim, num_heads=8, att_dropout=0.0, aropout=0.0) # cross_att(x=input_img, context=inputs, pad_mask=pad_mask) cross_att = CrossAttention(in_channels=3, emb_dim=emb_dim, att_dropout=0.0, aropout=0.0) cross_att(x=input_img, context=inputs, pad_mask=pad_mask)