戒酒的李白
24 lines
829 B
Python
24 lines
829 B
Python
import torch
|
|
import torch.nn as nn
|
|
|
|
class MultiHeadAttentionLayer(nn.Module):
|
|
def __init__(self, embed_size, num_heads):
|
|
super(MultiHeadAttentionLayer, self).__init__()
|
|
self.embed_size = embed_size
|
|
self.num_heads = num_heads
|
|
self.head_dim = embed_size // num_heads
|
|
|
|
assert (self.head_dim * num_heads == embed_size), "Embedding size needs to be divisible by num_heads"
|
|
|
|
# Define linear layers for Q, K, V
|
|
self.q_linear = nn.Linear(embed_size, embed_size)
|
|
self.k_linear = nn.Linear(embed_size, embed_size)
|
|
self.v_linear = nn.Linear(embed_size, embed_size)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
embed_size = 512
|
|
num_heads = 8
|
|
mha_layer = MultiHeadAttentionLayer(embed_size, num_heads)
|
|
print("Linear layers for Q, K, V initialized.")
|