Commit
·
fcb4463
1
Parent(s):
1c4fba2
Upload LUAR
Browse files- config.json +4 -1
- config.py +6 -0
- model.py +116 -6
config.json
CHANGED
@@ -7,7 +7,10 @@
|
|
7 |
"AutoModel": "model.LUAR"
|
8 |
},
|
9 |
"embedding_size": 512,
|
|
|
10 |
"model_type": "LUAR",
|
|
|
11 |
"torch_dtype": "float32",
|
12 |
-
"transformers_version": "4.33.2"
|
|
|
13 |
}
|
|
|
7 |
"AutoModel": "model.LUAR"
|
8 |
},
|
9 |
"embedding_size": 512,
|
10 |
+
"k_bucket_size": 1024,
|
11 |
"model_type": "LUAR",
|
12 |
+
"q_bucket_size": 512,
|
13 |
"torch_dtype": "float32",
|
14 |
+
"transformers_version": "4.33.2",
|
15 |
+
"use_memory_efficient_attention": false
|
16 |
}
|
config.py
CHANGED
@@ -6,7 +6,13 @@ class LUARConfig(PretrainedConfig):
|
|
6 |
|
7 |
def __init__(self,
|
8 |
embedding_size: int = 512,
|
|
|
|
|
|
|
9 |
**kwargs,
|
10 |
):
|
11 |
self.embedding_size = embedding_size
|
|
|
|
|
|
|
12 |
super().__init__(**kwargs)
|
|
|
6 |
|
7 |
def __init__(self,
|
8 |
embedding_size: int = 512,
|
9 |
+
use_memory_efficient_attention=False,
|
10 |
+
q_bucket_size=512,
|
11 |
+
k_bucket_size=1024,
|
12 |
**kwargs,
|
13 |
):
|
14 |
self.embedding_size = embedding_size
|
15 |
+
self.use_memory_efficient_attention = use_memory_efficient_attention
|
16 |
+
self.q_bucket_size = q_bucket_size
|
17 |
+
self.k_bucket_size = k_bucket_size
|
18 |
super().__init__(**kwargs)
|
model.py
CHANGED
@@ -1,29 +1,135 @@
|
|
1 |
|
2 |
import math
|
|
|
3 |
|
4 |
import torch
|
5 |
import torch.nn as nn
|
6 |
import torch.nn.functional as F
|
7 |
from einops import rearrange, reduce, repeat
|
|
|
8 |
from transformers import AutoModel, PreTrainedModel
|
9 |
|
10 |
from .config import LUARConfig
|
11 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
class SelfAttention(nn.Module):
|
13 |
"""Implements Dot-Product Self-Attention as used in "Attention is all You Need".
|
14 |
"""
|
15 |
-
def __init__(
|
|
|
|
|
|
|
|
|
|
|
16 |
super(SelfAttention, self).__init__()
|
|
|
|
|
|
|
17 |
|
18 |
def forward(self, k, q, v):
|
19 |
-
|
20 |
-
|
21 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
22 |
else:
|
23 |
d_k = q.size(-1)
|
24 |
scores = torch.matmul(k, q.transpose(-2, -1)) / math.sqrt(d_k)
|
25 |
p_attn = F.softmax(scores, dim=-1)
|
26 |
-
|
27 |
return torch.matmul(p_attn, v)
|
28 |
|
29 |
class LUAR(PreTrainedModel):
|
@@ -34,7 +140,11 @@ class LUAR(PreTrainedModel):
|
|
34 |
def __init__(self, config):
|
35 |
super().__init__(config)
|
36 |
self.create_transformer()
|
37 |
-
self.attn_fn = SelfAttention(
|
|
|
|
|
|
|
|
|
38 |
self.linear = nn.Linear(self.hidden_size, config.embedding_size)
|
39 |
|
40 |
def create_transformer(self):
|
|
|
1 |
|
2 |
import math
|
3 |
+
from functools import partial
|
4 |
|
5 |
import torch
|
6 |
import torch.nn as nn
|
7 |
import torch.nn.functional as F
|
8 |
from einops import rearrange, reduce, repeat
|
9 |
+
from torch.utils.checkpoint import checkpoint
|
10 |
from transformers import AutoModel, PreTrainedModel
|
11 |
|
12 |
from .config import LUARConfig
|
13 |
|
14 |
+
# Adapted LucidRains impl. of Memory Efficient Attention
|
15 |
+
# https://github.com/lucidrains/memory-efficient-attention-pytorch
|
16 |
+
|
17 |
+
def exists(val):
|
18 |
+
return val is not None
|
19 |
+
|
20 |
+
def summarize_qkv_chunk(
|
21 |
+
q, k, v,
|
22 |
+
mask
|
23 |
+
):
|
24 |
+
"""Dot-Product Attention for a chunk of queries, keys, and values.
|
25 |
+
"""
|
26 |
+
weight = torch.einsum('b h i d, b h j d -> b h i j', q, k)
|
27 |
+
|
28 |
+
if exists(mask):
|
29 |
+
# HuggingFace masks have to be added:
|
30 |
+
weight += mask
|
31 |
+
|
32 |
+
weight_max = weight.amax(dim = -1, keepdim = True).detach()
|
33 |
+
weight = weight - weight_max
|
34 |
+
|
35 |
+
exp_weight = weight.exp()
|
36 |
+
weighted_value = torch.einsum('b h i j, b h j d -> b h i d', exp_weight, v)
|
37 |
+
|
38 |
+
return exp_weight.sum(dim = -1), weighted_value, rearrange(weight_max, '... 1 -> ...')
|
39 |
+
|
40 |
+
checkpointed_summarize_qkv_chunk = partial(checkpoint, summarize_qkv_chunk)
|
41 |
+
|
42 |
+
def memory_efficient_attention(
|
43 |
+
q, k, v,
|
44 |
+
mask = None,
|
45 |
+
q_bucket_size = 512,
|
46 |
+
k_bucket_size = 1024,
|
47 |
+
eps = 1e-8
|
48 |
+
):
|
49 |
+
scale = q.shape[-1] ** -0.5
|
50 |
+
q = q * scale
|
51 |
+
|
52 |
+
# function
|
53 |
+
needs_backwards = q.requires_grad or k.requires_grad or v.requires_grad
|
54 |
+
summarize_qkv_fn = checkpointed_summarize_qkv_chunk if needs_backwards else summarize_qkv_chunk
|
55 |
+
|
56 |
+
# chunk all the inputs
|
57 |
+
q_chunks = q.split(q_bucket_size, dim = -2)
|
58 |
+
k_chunks = k.split(k_bucket_size, dim = -2)
|
59 |
+
v_chunks = v.split(k_bucket_size, dim = -2)
|
60 |
+
mask_chunks = mask.split(k_bucket_size, dim = -1) if exists(mask) else ((None,) * len(k_chunks))
|
61 |
+
|
62 |
+
# loop through all chunks and accumulate
|
63 |
+
out = []
|
64 |
+
for q_index, q_chunk in enumerate(q_chunks):
|
65 |
+
exp_weights = []
|
66 |
+
weighted_values = []
|
67 |
+
weight_maxes = []
|
68 |
+
|
69 |
+
for k_index, (k_chunk, v_chunk, mask_chunk) in enumerate(zip(k_chunks, v_chunks, mask_chunks)):
|
70 |
+
|
71 |
+
exp_weight_chunk, weighted_value_chunk, weight_max_chunk = summarize_qkv_fn(
|
72 |
+
q_chunk,
|
73 |
+
k_chunk,
|
74 |
+
v_chunk,
|
75 |
+
mask_chunk,
|
76 |
+
)
|
77 |
+
|
78 |
+
exp_weights.append(exp_weight_chunk)
|
79 |
+
weighted_values.append(weighted_value_chunk)
|
80 |
+
weight_maxes.append(weight_max_chunk)
|
81 |
+
|
82 |
+
exp_weights = torch.stack(exp_weights, dim = -1)
|
83 |
+
weighted_values = torch.stack(weighted_values, dim = -1)
|
84 |
+
weight_maxes = torch.stack(weight_maxes, dim = -1)
|
85 |
+
|
86 |
+
global_max = weight_maxes.amax(dim = -1, keepdim = True)
|
87 |
+
renorm_factor = (weight_maxes - global_max).exp().detach()
|
88 |
+
|
89 |
+
exp_weights = exp_weights * renorm_factor
|
90 |
+
weighted_values = weighted_values * rearrange(renorm_factor, '... c -> ... 1 c')
|
91 |
+
|
92 |
+
all_values = weighted_values.sum(dim = -1)
|
93 |
+
all_weights = exp_weights.sum(dim = -1)
|
94 |
+
|
95 |
+
normalized_values = all_values / (rearrange(all_weights, '... -> ... 1') + eps)
|
96 |
+
out.append(normalized_values)
|
97 |
+
|
98 |
+
return torch.cat(out, dim=-2)
|
99 |
+
|
100 |
class SelfAttention(nn.Module):
|
101 |
"""Implements Dot-Product Self-Attention as used in "Attention is all You Need".
|
102 |
"""
|
103 |
+
def __init__(
|
104 |
+
self,
|
105 |
+
memory_efficient_attention=False,
|
106 |
+
q_bucket_size=512,
|
107 |
+
k_bucket_size=1024,
|
108 |
+
):
|
109 |
super(SelfAttention, self).__init__()
|
110 |
+
self.use_memory_efficient_attention = memory_efficient_attention
|
111 |
+
self.q_bucket_size = q_bucket_size
|
112 |
+
self.k_bucket_size = k_bucket_size
|
113 |
|
114 |
def forward(self, k, q, v):
|
115 |
+
|
116 |
+
if self.use_memory_efficient_attention:
|
117 |
+
q, k, v = map(
|
118 |
+
lambda t: rearrange(t, 'b n (h d) -> b h n d', h = 12),
|
119 |
+
(q, k, v)
|
120 |
+
)
|
121 |
+
|
122 |
+
out = memory_efficient_attention(
|
123 |
+
q, k, v,
|
124 |
+
q_bucket_size=self.q_bucket_size,
|
125 |
+
k_bucket_size=self.k_bucket_size
|
126 |
+
)
|
127 |
+
out = rearrange(out, 'b h n d -> b n (h d)')
|
128 |
+
return out
|
129 |
else:
|
130 |
d_k = q.size(-1)
|
131 |
scores = torch.matmul(k, q.transpose(-2, -1)) / math.sqrt(d_k)
|
132 |
p_attn = F.softmax(scores, dim=-1)
|
|
|
133 |
return torch.matmul(p_attn, v)
|
134 |
|
135 |
class LUAR(PreTrainedModel):
|
|
|
140 |
def __init__(self, config):
|
141 |
super().__init__(config)
|
142 |
self.create_transformer()
|
143 |
+
self.attn_fn = SelfAttention(
|
144 |
+
config.use_memory_efficient_attention,
|
145 |
+
config.q_bucket_size,
|
146 |
+
config.k_bucket_size,
|
147 |
+
)
|
148 |
self.linear = nn.Linear(self.hidden_size, config.embedding_size)
|
149 |
|
150 |
def create_transformer(self):
|