File size: 10,259 Bytes
3b96cb1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
#    Copyright 2023 Haotian Liu
#
#    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 List, Optional, Union

import torch
import torch.nn as nn
from transformers import PreTrainedModel

DEFAULT_IMAGE_TOKEN = '<image>'
DEFAULT_IMAGE_PATCH_TOKEN = '<im_patch>'
DEFAULT_IM_START_TOKEN = '<im_start>'
DEFAULT_IM_END_TOKEN = '<im_end>'


class LlavaLlamaForCausalLM(PreTrainedModel):

    def __init__(self,
                 vision_encoder,
                 lang_encoder,
                 mm_hidden_size,
                 use_im_start_end=True,
                 use_mm_proj=True,
                 im_start_token: Optional[int] = None,
                 im_end_token: Optional[int] = None,
                 im_patch_token: Optional[int] = None,
                 mm_vision_select_layer: int = -1):
        super().__init__(lang_encoder.config)
        self.vision_tower = vision_encoder
        self.lang_encoder = lang_encoder

        self.use_im_start_end = use_im_start_end
        self.im_start_token = im_start_token
        self.im_end_token = im_end_token
        self.im_patch_token = im_patch_token
        self.mm_hidden_size = mm_hidden_size
        self.mm_vision_select_layer = mm_vision_select_layer
        self.lang_hidden_size = lang_encoder.config.hidden_size

        if use_mm_proj and not hasattr(lang_encoder.model, 'mm_projector'):
            mm_projector = nn.Linear(self.mm_hidden_size,
                                     self.lang_hidden_size)
            self.lang_encoder.model.add_module('mm_projector', mm_projector)
        elif not use_mm_proj:
            self.lang_encoder.model.add_module('mm_projector', nn.Identity())

        self.post_init()

    def forward(
        self,
        input_ids: torch.LongTensor = None,
        attention_mask: Optional[torch.Tensor] = None,
        past_key_values: Optional[List[torch.FloatTensor]] = None,
        inputs_embeds: Optional[torch.FloatTensor] = None,
        labels: Optional[torch.LongTensor] = None,
        use_cache: Optional[bool] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        images: Optional[torch.FloatTensor] = None,
        return_dict: Optional[bool] = None,
    ):
        output_attentions = (
            output_attentions if output_attentions is not None else
            self.config.output_attentions)
        output_hidden_states = (
            output_hidden_states if output_hidden_states is not None else
            self.config.output_hidden_states)
        return_dict = (
            return_dict
            if return_dict is not None else self.config.use_return_dict)

        # decoder outputs consists of
        # (dec_features, layer_state, dec_hidden, dec_attn)
        if inputs_embeds is None:
            inputs_embeds = self.lang_encoder.model.embed_tokens(input_ids)

        inputs_embeds = self.forward_vision_tower(input_ids, inputs_embeds,
                                                  images)

        return self.lang_encoder(
            input_ids=None,
            attention_mask=attention_mask,
            past_key_values=past_key_values,
            inputs_embeds=inputs_embeds,
            use_cache=use_cache,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
            labels=labels,
        )

    def prepare_inputs_for_generation(self,
                                      input_ids,
                                      past_key_values=None,
                                      attention_mask=None,
                                      inputs_embeds=None,
                                      **kwargs):
        if past_key_values:
            input_ids = input_ids[:, -1:]

        # if `inputs_embeds` are passed, we only want to use
        # them in the 1st generation step
        if inputs_embeds is not None and past_key_values is None:
            model_inputs = {'inputs_embeds': inputs_embeds}
        else:
            model_inputs = {'input_ids': input_ids}

        model_inputs.update({
            'past_key_values': past_key_values,
            'use_cache': kwargs.get('use_cache'),
            'attention_mask': attention_mask,
            'images': kwargs.get('images', None),
        })
        return model_inputs

    def forward_vision_tower(
        self,
        input_ids: torch.LongTensor,
        inputs_embeds: torch.FloatTensor,
        images: Union[torch.FloatTensor, list, None] = None,
    ):
        if self.use_im_start_end:
            assert self.im_start_token is not None
            assert self.im_end_token is not None
        if images is not None:
            assert self.im_patch_token is not None

        if self.vision_tower is None or images is None or (
                input_ids.shape[1] == 1 and not self.training):
            return inputs_embeds

        with torch.no_grad():
            if isinstance(images, (list, tuple)):
                # variable length images
                image_features = []
                for image in images:
                    feats = self.vision_tower(image.unsqueeze(0))
                    image_feature = feats[self.mm_vision_select_layer][:, 1:]
                    image_features.append(image_feature)
            else:
                feats = self.vision_tower(images)
                image_features = feats[self.mm_vision_select_layer][:, 1:]

        mm_projector = self.lang_encoder.model.mm_projector
        if isinstance(images, (list, tuple)):
            image_features = [
                mm_projector(image_feature)[0]
                for image_feature in image_features
            ]
        else:
            image_features = mm_projector(image_features)

        dummy_image_features = torch.zeros(
            256, 1024, device=inputs_embeds.device, dtype=inputs_embeds.dtype)
        dummy_image_features = mm_projector(dummy_image_features)

        new_input_embeds = []
        cur_image_idx = 0
        for cur_input_ids, cur_input_embeds in zip(input_ids, inputs_embeds):
            if (cur_input_ids != self.im_patch_token).all():
                # multimodal LLM, but the current sample is not multimodal
                cur_input_embeds = cur_input_embeds + (
                    0. * dummy_image_features).sum()
                new_input_embeds.append(cur_input_embeds)
                cur_image_idx += 1
                continue
            if self.use_im_start_end:
                cur_image_features = image_features[cur_image_idx]
                num_patches = cur_image_features.shape[0]
                if (cur_input_ids == self.im_start_token).sum() != (
                        cur_input_ids == self.im_end_token).sum():
                    raise ValueError('The number of image start tokens and '
                                     'image end tokens should be the same.')
                image_start_tokens = torch.where(
                    cur_input_ids == self.im_start_token)[0]
                for image_start_token_pos in image_start_tokens:
                    cur_image_features = image_features[cur_image_idx].to(
                        device=cur_input_embeds.device)
                    num_patches = cur_image_features.shape[0]
                    if cur_input_ids[image_start_token_pos + num_patches +
                                     1] != self.im_end_token:
                        raise ValueError('The image end token should follow '
                                         'the image start token.')
                    cur_new_input_embeds = torch.cat(
                        (cur_input_embeds[:image_start_token_pos + 1],
                         cur_image_features,
                         cur_input_embeds[image_start_token_pos + num_patches +
                                          1:]),
                        dim=0)
                    cur_image_idx += 1
                new_input_embeds.append(cur_new_input_embeds)
            else:
                cur_image_features = image_features[cur_image_idx]
                num_patches = cur_image_features.shape[0]
                if (cur_input_ids == self.im_patch_token).sum() != num_patches:
                    print(f'Debug: num_patches: {num_patches}')
                    raise ValueError(
                        'The number of image patch tokens should '
                        'be the same as the number of image patches.')
                masked_indices = torch.where(
                    cur_input_ids == self.im_patch_token)[0]
                mask_index_start = masked_indices[0]
                if (masked_indices != torch.arange(
                        mask_index_start,
                        mask_index_start + num_patches,
                        device=masked_indices.device,
                        dtype=masked_indices.dtype)).any():
                    raise ValueError(
                        'The image patch tokens should be consecutive.')
                cur_new_input_embeds = torch.cat(
                    (cur_input_embeds[:mask_index_start], cur_image_features,
                     cur_input_embeds[mask_index_start + num_patches:]),
                    dim=0)
                new_input_embeds.append(cur_new_input_embeds)
                cur_image_idx += 1
        inputs_embeds = torch.stack(new_input_embeds, dim=0)

        return inputs_embeds

    @staticmethod
    def _reorder_cache(past_key_values, beam_idx):
        reordered_past = ()
        for layer_past in past_key_values:
            reordered_past += (tuple(
                past_state.index_select(0, beam_idx)
                for past_state in layer_past), )
        return reordered_past