File size: 8,363 Bytes
52ca9c9
 
9763171
52ca9c9
 
 
 
9763171
52ca9c9
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9763171
52ca9c9
9763171
52ca9c9
 
 
 
 
 
9763171
52ca9c9
9763171
52ca9c9
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9763171
52ca9c9
 
9763171
52ca9c9
 
 
 
 
 
9763171
 
 
 
 
 
 
 
 
 
52ca9c9
 
 
9763171
52ca9c9
 
 
9763171
 
 
 
 
 
52ca9c9
 
 
 
 
 
 
 
9763171
 
52ca9c9
 
9763171
 
 
 
 
 
 
 
 
52ca9c9
 
 
 
 
 
 
 
 
9763171
52ca9c9
9763171
 
 
 
 
 
 
 
52ca9c9
 
9763171
52ca9c9
 
 
9763171
 
 
 
 
 
 
 
 
 
69b03ec
52ca9c9
 
 
 
 
 
 
 
 
 
 
 
 
9763171
 
52ca9c9
9763171
52ca9c9
9763171
d771171
52ca9c9
 
 
 
 
 
 
 
 
 
 
9763171
 
 
52ca9c9
 
 
 
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
import argparse
import time
from threading import Thread
from PIL import Image

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM, LlamaTokenizer
from transformers import StoppingCriteria, StoppingCriteriaList, TextIteratorStreamer

import dataclasses
from enum import auto, Enum
from typing import List, Tuple, Any

from minigpt4.common.registry import registry


class SeparatorStyle(Enum):
    """Different separator style."""
    SINGLE = auto()
    TWO = auto()


@dataclasses.dataclass
class Conversation:
    """A class that keeps all conversation history."""
    system: str
    roles: List[str]
    messages: List[List[str]]
    offset: int
    # system_img: List[Image.Image] = []
    sep_style: SeparatorStyle = SeparatorStyle.SINGLE
    sep: str = "###"
    sep2: str = None

    skip_next: bool = False
    conv_id: Any = None

    def get_prompt(self):
        if self.sep_style == SeparatorStyle.SINGLE:
            ret = self.system + self.sep
            for role, message in self.messages:
                if message:
                    ret += role + message + self.sep
                else:
                    ret += role
            return ret
        elif self.sep_style == SeparatorStyle.TWO:
            seps = [self.sep, self.sep2]
            ret = self.system + seps[0]
            for i, (role, message) in enumerate(self.messages):
                if message:
                    ret += role + message + seps[i % 2]
                else:
                    ret += role
            return ret
        else:
            raise ValueError(f"Invalid style: {self.sep_style}")

    def append_message(self, role, message):
        self.messages.append([role, message])

    def to_gradio_chatbot(self):
        ret = []
        for i, (role, msg) in enumerate(self.messages[self.offset:]):
            if i % 2 == 0:
                ret.append([msg, None])
            else:
                ret[-1][-1] = msg
        return ret

    def copy(self):
        return Conversation(
            system=self.system,
            # system_img=self.system_img,
            roles=self.roles,
            messages=[[x, y] for x, y in self.messages],
            offset=self.offset,
            sep_style=self.sep_style,
            sep=self.sep,
            sep2=self.sep2,
            conv_id=self.conv_id)

    def dict(self):
        return {
            "system": self.system,
            # "system_img": self.system_img,
            "roles": self.roles,
            "messages": self.messages,
            "offset": self.offset,
            "sep": self.sep,
            "sep2": self.sep2,
            "conv_id": self.conv_id,
        }


class StoppingCriteriaSub(StoppingCriteria):

    def __init__(self, stops=[], encounters=1):
        super().__init__()
        self.stops = stops

    def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTensor):
        for stop in self.stops:
            if torch.all((stop == input_ids[0][-len(stop):])).item():
                return True

        return False


CONV_VISION_Vicuna0 = Conversation(
    system="Give the following image: <Img>ImageContent</Img>. "
           "You will be able to see the image once I provide it to you. Please answer my questions.",
    roles=("Human: ", "Assistant: "),
    messages=[],
    offset=2,
    sep_style=SeparatorStyle.SINGLE,
    sep="###",
)

CONV_VISION_LLama2 = Conversation(
    system="Give the following image: <Img>ImageContent</Img>. "
           "You will be able to see the image once I provide it to you. Please answer my questions.",
    roles=("<s>[INST] ", " [/INST] "),
    messages=[],
    offset=2,
    sep_style=SeparatorStyle.SINGLE,
    sep="",
)



class Chat:
    def __init__(self, model, vis_processor, device='cuda:0', stopping_criteria=None):
        self.device = device
        self.model = model
        self.vis_processor = vis_processor

        if stopping_criteria is not None:
            self.stopping_criteria = stopping_criteria
        else:
            stop_words_ids = [torch.tensor([2]).to(self.device)]
            self.stopping_criteria = StoppingCriteriaList([StoppingCriteriaSub(stops=stop_words_ids)])

    def ask(self, text, conv):
        if len(conv.messages) > 0 and conv.messages[-1][0] == conv.roles[0] \
                and conv.messages[-1][1][-6:] == '</Img>':  # last message is image.
            conv.messages[-1][1] = ' '.join([conv.messages[-1][1], text])
        else:
            conv.append_message(conv.roles[0], text)

    def answer_prepare(self, conv, img_list, max_new_tokens=300, num_beams=1, min_length=1, top_p=0.9,
                       repetition_penalty=1.05, length_penalty=1, temperature=1.0, max_length=2000):
        conv.append_message(conv.roles[1], None)
        embs = self.get_context_emb(conv, img_list)

        current_max_len = embs.shape[1] + max_new_tokens
        if current_max_len - max_length > 0:
            print('Warning: The number of tokens in current conversation exceeds the max length. '
                  'The model will not see the contexts outside the range.')
        begin_idx = max(0, current_max_len - max_length)
        embs = embs[:, begin_idx:]

        generation_kwargs = dict(
            inputs_embeds=embs,
            max_new_tokens=max_new_tokens,
            stopping_criteria=self.stopping_criteria,
            num_beams=num_beams,
            do_sample=True,
            min_length=min_length,
            top_p=top_p,
            repetition_penalty=repetition_penalty,
            length_penalty=length_penalty,
            temperature=float(temperature),
        )
        return generation_kwargs

    def answer(self, conv, img_list, **kargs):
        generation_dict = self.answer_prepare(conv, img_list, **kargs)

        output_token = self.model.llama_model.generate(**generation_dict)[0]
        output_text = self.model.llama_tokenizer.decode(output_token, skip_special_tokens=True)

        output_text = output_text.split('###')[0]  # remove the stop sign '###'
        output_text = output_text.split('Assistant:')[-1].strip()

        conv.messages[-1][1] = output_text
        return output_text, output_token.cpu().numpy()

    def stream_answer(self, conv, img_list, **kargs):
        generation_kwargs = self.answer_prepare(conv, img_list, **kargs)
        streamer = TextIteratorStreamer(self.model.llama_tokenizer, skip_special_tokens=True)
        generation_kwargs['streamer'] = streamer
        thread = Thread(target=self.model.llama_model.generate, kwargs=generation_kwargs)
        thread.start()
        return streamer

    def encode_img(self, img_list):
        image = img_list[0]
        img_list.pop(0)
        if isinstance(image, str):  # is a image path
            raw_image = Image.open(image).convert('RGB')
            image = self.vis_processor(raw_image).unsqueeze(0).to(self.device)
        elif isinstance(image, Image.Image):
            raw_image = image
            image = self.vis_processor(raw_image).unsqueeze(0).to(self.device)
        elif isinstance(image, torch.Tensor):
            if len(image.shape) == 3:
                image = image.unsqueeze(0)
            image = image.to(self.device)

        image_emb, _ = self.model.encode_img(image)
        img_list.append(image_emb)

    def upload_img(self, image, conv, img_list):
        conv.append_message(conv.roles[0], "<Img><ImageHere></Img>")
        img_list.append(image)
        msg = "Received."

        return msg

    def get_context_emb(self, conv, img_list):
        prompt = conv.get_prompt()
        prompt_segs = prompt.split('<ImageHere>')
        assert len(prompt_segs) == len(img_list) + 1, "Unmatched numbers of image placeholders and images."
        seg_tokens = [
            self.model.llama_tokenizer(
                seg, return_tensors="pt", add_special_tokens=i == 0).to(self.device).input_ids
            # only add bos to the first seg
            for i, seg in enumerate(prompt_segs)
        ]
        print('debug device: ', self.device)
        print('debug model device: ', self.model.device)
        seg_embs = [self.model.embed_tokens(seg_t) for seg_t in seg_tokens]
        mixed_embs = [emb for pair in zip(seg_embs[:-1], img_list) for emb in pair] + [seg_embs[-1]]
        mixed_embs = torch.cat(mixed_embs, dim=1)
        return mixed_embs