File size: 6,549 Bytes
8b54513
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
from typing import List
import torch
import torch.nn as nn
import json
import os
from .tokenizer import Tokenizer
from . import LLM

from fairscale.nn.model_parallel import initialize as fs_init


class MetaModel(nn.Module):

    def __init__(self, llama_type, llama_config, llama_ckpt_dir=None, tokenizer_path=None):
        super().__init__()

        self.criterion = torch.nn.CrossEntropyLoss(ignore_index=0)

        ModelArgs = LLM.__dict__[llama_type].ModelArgs
        Transformer = LLM.__dict__[llama_type].Transformer

        with open(llama_config, "r") as f:
            params = json.loads(f.read())
        model_args: ModelArgs = ModelArgs(
            max_seq_len=2048, max_batch_size=32, **params
        )
        self.tokenizer = Tokenizer(model_path=tokenizer_path)
        model_args.vocab_size = self.tokenizer.n_words

        model = Transformer(model_args)
        mp_rank = fs_init.get_model_parallel_rank()
        if llama_ckpt_dir is not None:
            ckpt_path = os.path.join(llama_ckpt_dir, f"consolidated.{mp_rank:02d}.pth")
            if os.path.exists(ckpt_path):
                checkpoint = torch.load(ckpt_path, map_location="cpu")
                msg = model.load_state_dict(checkpoint, strict=False)
                print(msg)
            else:
                print(f'Checkpoint not found at {ckpt_path}')
        self.llma = model
        for name, param in self.named_parameters():
            if param.requires_grad:
               print(f"Trainable param: {name}, {param.shape}, {param.dtype}")
        count = sum(p.numel() for p in self.parameters() if p.requires_grad)
        print(f"Parameter count : {count}")

    def forward(self, examples, labels, image=None, modal='image'):
        output = self.llma(examples, image=image, modal=modal)
        output = output[:, :-1, :]
        labels = labels[:, 1:]

        if labels.sum() == 0:
            c_loss = output.mean() * 0
        else:
            c_loss = self.criterion(output.reshape(-1, 32000), labels.flatten())

        return c_loss

    def generate(
        self,
        prompts: List[str],
        images,
        max_gen_len: int,
        temperature: float = 0.8,
        top_p: float = 0.95,
        modal = ['image'],
    ) -> List[str]:
        bsz = len(prompts)
        params = self.llma.params
        assert bsz <= params.max_batch_size, (bsz, params.max_batch_size)

        prompt_tokens = [self.tokenizer.encode(
            x, bos=True, eos=False) for x in prompts]

        min_prompt_size = min([len(t) for t in prompt_tokens])
        max_prompt_size = max([len(t) for t in prompt_tokens])

        total_len = min(params.max_seq_len, max_gen_len + max_prompt_size)

        tokens = torch.full(
            (bsz, total_len), self.tokenizer.pad_id).cuda().long()
        for k, t in enumerate(prompt_tokens):
            tokens[k, : len(t)] = torch.tensor(t).long()
        input_text_mask = tokens != self.tokenizer.pad_id
        start_pos = min_prompt_size
        prev_pos = 0
        for cur_pos in range(start_pos, total_len):
            logits = self.llma.forward_inference(tokens[:, prev_pos:cur_pos], prev_pos, images if prev_pos == 0 else None, modal=modal)
            if temperature > 0:
                probs = torch.softmax(logits / temperature, dim=-1)
                next_token = self.sample_top_p(probs, top_p)
            else:
                next_token = torch.argmax(logits, dim=-1)
            next_token = next_token.reshape(-1)
            # only replace token if prompt has already been generated
            next_token = torch.where(
                input_text_mask[:, cur_pos], tokens[:, cur_pos], next_token
            )
            tokens[:, cur_pos] = next_token
            prev_pos = cur_pos

        decoded = []
        for i, t in enumerate(tokens.tolist()):
            # cut to max gen len
            t = t[: len(prompt_tokens[i]) + max_gen_len]
            # cut to eos tok if any
            try:
                t = t[: t.index(self.tokenizer.eos_id)]
            except ValueError:
                pass
            decoded.append(self.tokenizer.decode(t))
        return decoded
    
    @torch.inference_mode()
    def stream_generate(
        self,
        prompt: str,
        images,
        max_gen_len: int,
        temperature: float = 0.8,
        top_p: float = 0.95,
        modal = ['image'],
    ):
        params = self.llma.params

        prompt_tokens = self.tokenizer.encode(prompt, bos=True, eos=False)
        # truncate from the left. leave some space for generation.
        max_seq_len = params.max_seq_len
        if images is not None:
            max_seq_len -= self.llma.image_words

        max_prompt_size = max_seq_len - max_gen_len
        prompt_tokens = prompt_tokens[-max_prompt_size:]

        prompt_size = len(prompt_tokens)

        total_len = min(max_seq_len, max_gen_len + prompt_size)

        tokens = torch.full([total_len], 0).cuda().long()

        tokens[:len(prompt_tokens)] = torch.tensor(prompt_tokens).long()
        start_pos = prompt_size
        prev_pos = 0
        generate_until = start_pos
        for cur_pos in range(start_pos, total_len):
            logits = self.llma.forward_inference(tokens[None, prev_pos:cur_pos], prev_pos, images if prev_pos == 0 else None, modal = modal)
            if temperature > 0:
                probs = torch.softmax(logits / temperature, dim=-1)
                next_token = self.sample_top_p(probs, top_p)
            else:
                next_token = torch.argmax(logits, dim=-1)
            next_token = next_token.item()

            if next_token == self.tokenizer.eos_id:
                break

            tokens[cur_pos] = next_token
            prev_pos = cur_pos
            generate_until = cur_pos + 1
            yield {"text": self.tokenizer.decode(tokens[start_pos:generate_until].tolist()), "end_of_content": False}

        yield {"text": self.tokenizer.decode(tokens[start_pos:generate_until].tolist()), "end_of_content": True}

    def sample_top_p(self, probs, p):
        probs_sort, probs_idx = torch.sort(probs, dim=-1, descending=True)
        probs_sum = torch.cumsum(probs_sort, dim=-1)
        mask = probs_sum - probs_sort > p
        probs_sort[mask] = 0.0
        probs_sort.div_(probs_sort.sum(dim=-1, keepdim=True))
        next_token = torch.multinomial(probs_sort, num_samples=1)
        next_token = torch.gather(probs_idx, -1, next_token)
        return next_token

    def get_image_words(self):
        return self.llma.image_words