Skip to content

vllm.transformers_utils.configs.minimax_vl_01

MiniMaxVL01 model configuration

MiniMaxVL01Config

Bases: PretrainedConfig

Source code in vllm/transformers_utils/configs/minimax_vl_01.py 
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
class MiniMaxVL01Config(PretrainedConfig):
    model_type = "minimax_vl_01"

    def __init__(
        self,
        vision_config=None,
        text_config=None,
        ignore_index=-100,
        image_token_index=32000,
        projector_hidden_act="gelu",
        vision_feature_select_strategy="default",
        vision_feature_layer=-2,
        image_grid_pinpoints=None,
        tie_word_embeddings=False,
        image_seq_length=576,
        **kwargs,
    ):
        self.ignore_index = ignore_index
        self.image_token_index = image_token_index
        self.projector_hidden_act = projector_hidden_act
        self.image_seq_length = image_seq_length

        if vision_feature_select_strategy not in ["default", "full"]:
            raise ValueError("vision_feature_select_strategy should " +
                             "be one of 'default', 'full'." +
                             f"Got: {vision_feature_select_strategy}")

        self.vision_feature_select_strategy = vision_feature_select_strategy
        self.vision_feature_layer = vision_feature_layer
        image_grid_pinpoints = (
            image_grid_pinpoints if image_grid_pinpoints is not None else
            [[336, 672], [672, 336], [672, 672], [1008, 336], [336, 1008]])
        self.image_grid_pinpoints = image_grid_pinpoints

        if isinstance(vision_config, dict):
            if "model_type" not in vision_config:
                vision_config["model_type"] = "clip_vision_model"
            vision_config = CONFIG_MAPPING[vision_config["model_type"]](
                **vision_config)
        elif vision_config is None:
            vision_config = CONFIG_MAPPING["clip_vision_model"](
                intermediate_size=4096,
                hidden_size=1024,
                patch_size=14,
                image_size=336,
                num_hidden_layers=24,
                num_attention_heads=16,
                vocab_size=32000,
                projection_dim=768,
            )

        self.vision_config = vision_config

        if text_config is not None:
            text_config = MiniMaxText01Config(**text_config)
        else:
            text_config = MiniMaxText01Config()

        self.text_config = text_config

        super().__init__(tie_word_embeddings=tie_word_embeddings, **kwargs)

ignore_index instance-attribute 

ignore_index = ignore_index

image_grid_pinpoints instance-attribute 

image_grid_pinpoints = image_grid_pinpoints

image_seq_length instance-attribute 

image_seq_length = image_seq_length

image_token_index instance-attribute 

image_token_index = image_token_index

model_type class-attribute instance-attribute 

model_type = 'minimax_vl_01'

projector_hidden_act instance-attribute 

projector_hidden_act = projector_hidden_act

text_config instance-attribute 

text_config = text_config

vision_config instance-attribute 

vision_config = vision_config

vision_feature_layer instance-attribute 

vision_feature_layer = vision_feature_layer

vision_feature_select_strategy instance-attribute 

vision_feature_select_strategy = (
    vision_feature_select_strategy
)

__init__ 

__init__(
    vision_config=None,
    text_config=None,
    ignore_index=-100,
    image_token_index=32000,
    projector_hidden_act="gelu",
    vision_feature_select_strategy="default",
    vision_feature_layer=-2,
    image_grid_pinpoints=None,
    tie_word_embeddings=False,
    image_seq_length=576,
    **kwargs,
)
Source code in vllm/transformers_utils/configs/minimax_vl_01.py 
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
def __init__(
    self,
    vision_config=None,
    text_config=None,
    ignore_index=-100,
    image_token_index=32000,
    projector_hidden_act="gelu",
    vision_feature_select_strategy="default",
    vision_feature_layer=-2,
    image_grid_pinpoints=None,
    tie_word_embeddings=False,
    image_seq_length=576,
    **kwargs,
):
    self.ignore_index = ignore_index
    self.image_token_index = image_token_index
    self.projector_hidden_act = projector_hidden_act
    self.image_seq_length = image_seq_length

    if vision_feature_select_strategy not in ["default", "full"]:
        raise ValueError("vision_feature_select_strategy should " +
                         "be one of 'default', 'full'." +
                         f"Got: {vision_feature_select_strategy}")

    self.vision_feature_select_strategy = vision_feature_select_strategy
    self.vision_feature_layer = vision_feature_layer
    image_grid_pinpoints = (
        image_grid_pinpoints if image_grid_pinpoints is not None else
        [[336, 672], [672, 336], [672, 672], [1008, 336], [336, 1008]])
    self.image_grid_pinpoints = image_grid_pinpoints

    if isinstance(vision_config, dict):
        if "model_type" not in vision_config:
            vision_config["model_type"] = "clip_vision_model"
        vision_config = CONFIG_MAPPING[vision_config["model_type"]](
            **vision_config)
    elif vision_config is None:
        vision_config = CONFIG_MAPPING["clip_vision_model"](
            intermediate_size=4096,
            hidden_size=1024,
            patch_size=14,
            image_size=336,
            num_hidden_layers=24,
            num_attention_heads=16,
            vocab_size=32000,
            projection_dim=768,
        )

    self.vision_config = vision_config

    if text_config is not None:
        text_config = MiniMaxText01Config(**text_config)
    else:
        text_config = MiniMaxText01Config()

    self.text_config = text_config

    super().__init__(tie_word_embeddings=tie_word_embeddings, **kwargs)