File size: 17,640 Bytes
21231ee
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
# Copyright 2022 The HuggingFace Team. All rights reserved.
#
# 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
#
#     http://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.
from typing import Optional, Union

import paddle
import paddle.nn as nn
import paddle.nn.functional as F

from ..initializer import normal_, zeros_


class CrossAttention(nn.Layer):
    r"""
    A cross attention layer.

    Parameters:
        query_dim (`int`): The number of channels in the query.
        cross_attention_dim (`int`, *optional*):
            The number of channels in the encoder_hidden_states. If not given, defaults to `query_dim`.
        heads (`int`,  *optional*, defaults to 8): The number of heads to use for multi-head attention.
        dim_head (`int`,  *optional*, defaults to 64): The number of channels in each head.
        dropout (`float`, *optional*, defaults to 0.0): The dropout probability to use.
        bias (`bool`, *optional*, defaults to False):
            Set to `True` for the query, key, and value linear layers to contain a bias parameter.
    """

    def __init__(
        self,
        query_dim: int,
        cross_attention_dim: Optional[int] = None,
        heads: int = 8,
        dim_head: int = 64,
        dropout: float = 0.0,
        bias=False,
        upcast_attention: bool = False,
        upcast_softmax: bool = False,
        added_kv_proj_dim: Optional[int] = None,
        norm_num_groups: Optional[int] = None,
        processor: Optional["AttnProcessor"] = None,
    ):
        super().__init__()
        inner_dim = dim_head * heads
        cross_attention_dim = cross_attention_dim if cross_attention_dim is not None else query_dim
        self.upcast_attention = upcast_attention
        self.upcast_softmax = upcast_softmax

        self.scale = dim_head**-0.5
        self.num_heads = heads
        self.head_dim = inner_dim // heads
        # for slice_size > 0 the attention score computation
        # is split across the batch axis to save memory
        # You can set slice_size with `set_attention_slice`
        self.sliceable_head_dim = heads

        self.added_kv_proj_dim = added_kv_proj_dim

        if norm_num_groups is not None:
            self.group_norm = nn.GroupNorm(num_channels=inner_dim, num_groups=norm_num_groups, epsilon=1e-5)
        else:
            self.group_norm = None

        self.to_q = nn.Linear(query_dim, inner_dim, bias_attr=bias)
        self.to_k = nn.Linear(cross_attention_dim, inner_dim, bias_attr=bias)
        self.to_v = nn.Linear(cross_attention_dim, inner_dim, bias_attr=bias)

        if self.added_kv_proj_dim is not None:
            self.add_k_proj = nn.Linear(added_kv_proj_dim, cross_attention_dim)
            self.add_v_proj = nn.Linear(added_kv_proj_dim, cross_attention_dim)

        self.to_out = nn.LayerList([])
        self.to_out.append(nn.Linear(inner_dim, query_dim))
        self.to_out.append(nn.Dropout(dropout))

        # set attention processor
        processor = processor if processor is not None else CrossAttnProcessor()
        self.set_processor(processor)

    def set_attention_slice(self, slice_size):
        if slice_size is not None and slice_size > self.sliceable_head_dim:
            raise ValueError(f"slice_size {slice_size} has to be smaller or equal to {self.sliceable_head_dim}.")

        if slice_size is not None and self.added_kv_proj_dim is not None:
            processor = SlicedAttnAddedKVProcessor(slice_size)
        elif slice_size is not None:
            processor = SlicedAttnProcessor(slice_size)
        elif self.added_kv_proj_dim is not None:
            processor = CrossAttnAddedKVProcessor()
        else:
            processor = CrossAttnProcessor()

        self.set_processor(processor)

    def set_processor(self, processor: "AttnProcessor"):
        self.processor = processor

    def forward(self, hidden_states, encoder_hidden_states=None, attention_mask=None, **cross_attention_kwargs):
        # The `CrossAttention` class can call different attention processors / attention functions
        # here we simply pass along all tensors to the selected processor class
        # For standard processors that are defined here, `**cross_attention_kwargs` is empty
        return self.processor(
            self,
            hidden_states,
            encoder_hidden_states=encoder_hidden_states,
            attention_mask=attention_mask,
            **cross_attention_kwargs,
        )

    def batch_to_head_dim(self, tensor):
        tensor = tensor.transpose([0, 2, 1, 3])
        tensor = tensor.reshape([0, 0, tensor.shape[2] * tensor.shape[3]])
        return tensor

    def head_to_batch_dim(self, tensor):
        tensor = tensor.reshape([0, 0, self.num_heads, self.head_dim])
        tensor = tensor.transpose([0, 2, 1, 3])
        return tensor

    def get_attention_scores(self, query, key, attention_mask=None):
        if self.upcast_attention:
            query = query.cast("float32")
            key = key.cast("float32")

        attention_scores = paddle.matmul(query, key, transpose_y=True) * self.scale

        if attention_mask is not None:
            attention_scores = attention_scores + attention_mask

        if self.upcast_softmax:
            attention_scores = attention_scores.cast("float32")

        attention_probs = F.softmax(attention_scores, axis=-1)
        if self.upcast_softmax:
            attention_probs = attention_probs.cast(query.dtype)

        return attention_probs

    def prepare_attention_mask(self, attention_mask, target_length):
        if attention_mask is None:
            return attention_mask

        if attention_mask.shape[-1] != target_length:
            attention_mask = F.pad(attention_mask, (0, target_length), value=0.0, data_format="NCL")
            attention_mask = attention_mask.repeat_interleave(self.num_heads, axis=0)
        return attention_mask


class CrossAttnProcessor:
    def __call__(self, attn: CrossAttention, hidden_states, encoder_hidden_states=None, attention_mask=None):
        batch_size, sequence_length, _ = hidden_states.shape
        attention_mask = attn.prepare_attention_mask(attention_mask, sequence_length)
        attention_mask = (
            attention_mask.reshape([batch_size, attn.num_heads, -1, attention_mask.shape[-1]])
            if attention_mask is not None
            else None
        )

        query = attn.to_q(hidden_states)
        query = attn.head_to_batch_dim(query)

        encoder_hidden_states = encoder_hidden_states if encoder_hidden_states is not None else hidden_states
        key = attn.to_k(encoder_hidden_states)
        value = attn.to_v(encoder_hidden_states)
        key = attn.head_to_batch_dim(key)
        value = attn.head_to_batch_dim(value)

        attention_probs = attn.get_attention_scores(query, key, attention_mask)
        hidden_states = paddle.matmul(attention_probs, value)
        hidden_states = attn.batch_to_head_dim(hidden_states)

        # linear proj
        hidden_states = attn.to_out[0](hidden_states)
        # dropout
        hidden_states = attn.to_out[1](hidden_states)

        return hidden_states


class LoRALinearLayer(nn.Layer):
    def __init__(self, in_features, out_features, rank=4):
        super().__init__()

        if rank > min(in_features, out_features):
            raise ValueError(f"LoRA rank {rank} must be less or equal than {min(in_features, out_features)}")

        self.down = nn.Linear(in_features, rank, bias_attr=False)
        self.up = nn.Linear(rank, out_features, bias_attr=False)
        self.scale = 1.0

        normal_(self.down.weight, std=1 / rank)
        zeros_(self.up.weight)

    def forward(self, hidden_states):
        orig_dtype = hidden_states.dtype
        dtype = self.down.weight.dtype

        down_hidden_states = self.down(hidden_states.cast(dtype))
        up_hidden_states = self.up(down_hidden_states)

        return up_hidden_states.cast(orig_dtype)


class LoRACrossAttnProcessor(nn.Layer):
    def __init__(self, hidden_size, cross_attention_dim=None, rank=4):
        super().__init__()

        self.hidden_size = hidden_size
        self.cross_attention_dim = cross_attention_dim
        self.rank = rank

        self.to_q_lora = LoRALinearLayer(hidden_size, hidden_size, rank)
        self.to_k_lora = LoRALinearLayer(cross_attention_dim or hidden_size, hidden_size, rank)
        self.to_v_lora = LoRALinearLayer(cross_attention_dim or hidden_size, hidden_size, rank)
        self.to_out_lora = LoRALinearLayer(hidden_size, hidden_size, rank)

    def __call__(
        self, attn: CrossAttention, hidden_states, encoder_hidden_states=None, attention_mask=None, scale=1.0
    ):
        batch_size, sequence_length, _ = hidden_states.shape
        attention_mask = attn.prepare_attention_mask(attention_mask, sequence_length)
        attention_mask = (
            attention_mask.reshape([batch_size, attn.num_heads, -1, attention_mask.shape[-1]])
            if attention_mask is not None
            else None
        )

        query = attn.to_q(hidden_states) + scale * self.to_q_lora(hidden_states)
        query = attn.head_to_batch_dim(query)

        encoder_hidden_states = encoder_hidden_states if encoder_hidden_states is not None else hidden_states

        key = attn.to_k(encoder_hidden_states) + scale * self.to_k_lora(encoder_hidden_states)
        value = attn.to_v(encoder_hidden_states) + scale * self.to_v_lora(encoder_hidden_states)

        key = attn.head_to_batch_dim(key)
        value = attn.head_to_batch_dim(value)

        attention_probs = attn.get_attention_scores(query, key, attention_mask)
        hidden_states = paddle.matmul(attention_probs, value)
        hidden_states = attn.batch_to_head_dim(hidden_states)

        # linear proj
        hidden_states = attn.to_out[0](hidden_states) + scale * self.to_out_lora(hidden_states)
        # dropout
        hidden_states = attn.to_out[1](hidden_states)

        return hidden_states


class CrossAttnAddedKVProcessor:
    def __call__(self, attn: CrossAttention, hidden_states, encoder_hidden_states=None, attention_mask=None):
        residual = hidden_states
        hidden_states = hidden_states.reshape([hidden_states.shape[0], hidden_states.shape[1], -1]).transpose(
            [0, 2, 1]
        )
        batch_size, sequence_length, _ = hidden_states.shape
        encoder_hidden_states = encoder_hidden_states.transpose([0, 2, 1])

        attention_mask = attn.prepare_attention_mask(attention_mask, sequence_length)
        attention_mask = (
            attention_mask.reshape([batch_size, attn.num_heads, -1, attention_mask.shape[-1]])
            if attention_mask is not None
            else None
        )

        hidden_states = attn.group_norm(hidden_states.transpose([0, 2, 1])).transpose([0, 2, 1])

        query = attn.to_q(hidden_states)
        query = attn.head_to_batch_dim(query)

        key = attn.to_k(hidden_states)
        value = attn.to_v(hidden_states)
        key = attn.head_to_batch_dim(key)
        value = attn.head_to_batch_dim(value)

        encoder_hidden_states_key_proj = attn.add_k_proj(encoder_hidden_states)
        encoder_hidden_states_value_proj = attn.add_v_proj(encoder_hidden_states)
        encoder_hidden_states_key_proj = attn.head_to_batch_dim(encoder_hidden_states_key_proj)
        encoder_hidden_states_value_proj = attn.head_to_batch_dim(encoder_hidden_states_value_proj)

        key = paddle.concat([encoder_hidden_states_key_proj, key], axis=2)
        value = paddle.concat([encoder_hidden_states_value_proj, value], axis=2)

        attention_probs = attn.get_attention_scores(query, key, attention_mask)
        hidden_states = paddle.matmul(attention_probs, value)
        hidden_states = attn.batch_to_head_dim(hidden_states)

        # linear proj
        hidden_states = attn.to_out[0](hidden_states)
        # dropout
        hidden_states = attn.to_out[1](hidden_states)

        hidden_states = hidden_states.transpose([0, 2, 1]).reshape(residual.shape)
        hidden_states = hidden_states + residual

        return hidden_states


class SlicedAttnProcessor:
    def __init__(self, slice_size):
        self.slice_size = slice_size

    def __call__(self, attn: CrossAttention, hidden_states, encoder_hidden_states=None, attention_mask=None):
        batch_size, sequence_length, _ = hidden_states.shape

        attention_mask = attn.prepare_attention_mask(attention_mask, sequence_length)

        query = attn.to_q(hidden_states)
        query = attn.head_to_batch_dim(query)

        encoder_hidden_states = encoder_hidden_states if encoder_hidden_states is not None else hidden_states
        key = attn.to_k(encoder_hidden_states)
        value = attn.to_v(encoder_hidden_states)
        key = attn.head_to_batch_dim(key)
        value = attn.head_to_batch_dim(value)

        query = query.flatten(0, 1)
        key = key.flatten(0, 1)
        value = value.flatten(0, 1)

        batch_size_attention = query.shape[0]
        hidden_states = paddle.zeros((batch_size_attention, sequence_length, attn.head_dim), dtype=query.dtype)

        for i in range(hidden_states.shape[0] // self.slice_size):
            start_idx = i * self.slice_size
            end_idx = (i + 1) * self.slice_size

            query_slice = query[start_idx:end_idx]
            key_slice = key[start_idx:end_idx]
            attn_mask_slice = attention_mask[start_idx:end_idx] if attention_mask is not None else None

            attn_slice = attn.get_attention_scores(query_slice, key_slice, attn_mask_slice)

            attn_slice = paddle.matmul(attn_slice, value[start_idx:end_idx])

            hidden_states[start_idx:end_idx] = attn_slice

        # reshape back to [bs, num_heads, seqlen, head_dim]
        hidden_states = hidden_states.reshape([-1, attn.num_heads, sequence_length, attn.head_dim])
        # reshape hidden_states
        hidden_states = attn.batch_to_head_dim(hidden_states)

        # linear proj
        hidden_states = attn.to_out[0](hidden_states)
        # dropout
        hidden_states = attn.to_out[1](hidden_states)

        return hidden_states


class SlicedAttnAddedKVProcessor:
    def __init__(self, slice_size):
        self.slice_size = slice_size

    def __call__(self, attn: "CrossAttention", hidden_states, encoder_hidden_states=None, attention_mask=None):
        residual = hidden_states
        hidden_states = hidden_states.reshape([hidden_states.shape[0], hidden_states.shape[1], -1]).transpose(
            [0, 2, 1]
        )
        encoder_hidden_states = encoder_hidden_states.transpose([0, 2, 1])

        batch_size, sequence_length, _ = hidden_states.shape

        attention_mask = attn.prepare_attention_mask(attention_mask, sequence_length)

        hidden_states = attn.group_norm(hidden_states.transpose([0, 2, 1])).transpose([0, 2, 1])

        query = attn.to_q(hidden_states)
        query = attn.head_to_batch_dim(query)

        key = attn.to_k(hidden_states)
        value = attn.to_v(hidden_states)
        encoder_hidden_states_key_proj = attn.add_k_proj(encoder_hidden_states)
        encoder_hidden_states_value_proj = attn.add_v_proj(encoder_hidden_states)

        key = attn.head_to_batch_dim(key)
        value = attn.head_to_batch_dim(value)
        encoder_hidden_states_key_proj = attn.head_to_batch_dim(encoder_hidden_states_key_proj)
        encoder_hidden_states_value_proj = attn.head_to_batch_dim(encoder_hidden_states_value_proj)

        key = paddle.concat([encoder_hidden_states_key_proj, key], axis=2)
        value = paddle.concat([encoder_hidden_states_value_proj, value], axis=2)

        query = query.flatten(0, 1)
        key = key.flatten(0, 1)
        value = value.flatten(0, 1)

        batch_size_attention = query.shape[0]
        hidden_states = paddle.zeros((batch_size_attention, sequence_length, attn.head_dim), dtype=query.dtype)
        for i in range(hidden_states.shape[0] // self.slice_size):
            start_idx = i * self.slice_size
            end_idx = (i + 1) * self.slice_size

            query_slice = query[start_idx:end_idx]
            key_slice = key[start_idx:end_idx]
            attn_mask_slice = attention_mask[start_idx:end_idx] if attention_mask is not None else None

            attn_slice = attn.get_attention_scores(query_slice, key_slice, attn_mask_slice)

            attn_slice = paddle.matmul(attn_slice, value[start_idx:end_idx])

            hidden_states[start_idx:end_idx] = attn_slice

        # reshape back to [bs, num_heads, seqlen, head_dim]
        hidden_states = hidden_states.reshape([-1, attn.num_heads, sequence_length, attn.head_dim])
        # reshape hidden_states
        hidden_states = attn.batch_to_head_dim(hidden_states)

        # linear proj
        hidden_states = attn.to_out[0](hidden_states)
        # dropout
        hidden_states = attn.to_out[1](hidden_states)

        hidden_states = hidden_states.transpose([0, 2, 1]).reshape(residual.shape)
        hidden_states = hidden_states + residual

        return hidden_states


AttnProcessor = Union[
    CrossAttnProcessor,
    SlicedAttnProcessor,
    CrossAttnAddedKVProcessor,
    SlicedAttnAddedKVProcessor,
]