Skip to content

vllm.entrypoints.pooling.classify.protocol

ClassificationRequest module-attribute 

ClassificationRequest: TypeAlias = (
    ClassificationCompletionRequest
    | ClassificationChatRequest
)

ClassificationChatRequest

Bases: PoolingBasicRequestMixin, ChatRequestMixin

Source code in vllm/entrypoints/pooling/classify/protocol.py 
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
class ClassificationChatRequest(PoolingBasicRequestMixin, ChatRequestMixin):
    # --8<-- [start:chat-classification-extra-params]
    mm_processor_kwargs: dict[str, Any] | None = Field(
        default=None,
        description=("Additional kwargs to pass to the HF processor."),
    )

    softmax: bool | None = Field(
        default=None,
        description="softmax will be deprecated, please use use_activation instead.",
    )

    activation: bool | None = Field(
        default=None,
        description="activation will be deprecated, please use use_activation instead.",
    )

    use_activation: bool | None = Field(
        default=None,
        description="Whether to use activation for classification outputs. "
        "Default is True.",
    )
    # --8<-- [end:chat-classification-extra-params]

    def to_pooling_params(self):
        return PoolingParams(
            truncate_prompt_tokens=self.truncate_prompt_tokens,
            use_activation=get_use_activation(self),
        )

activation class-attribute instance-attribute 

activation: bool | None = Field(
    default=None,
    description="activation will be deprecated, please use use_activation instead.",
)

mm_processor_kwargs class-attribute instance-attribute 

mm_processor_kwargs: dict[str, Any] | None = Field(
    default=None,
    description="Additional kwargs to pass to the HF processor.",
)

softmax class-attribute instance-attribute 

softmax: bool | None = Field(
    default=None,
    description="softmax will be deprecated, please use use_activation instead.",
)

use_activation class-attribute instance-attribute 

use_activation: bool | None = Field(
    default=None,
    description="Whether to use activation for classification outputs. Default is True.",
)

to_pooling_params 

to_pooling_params()
Source code in vllm/entrypoints/pooling/classify/protocol.py 
72
73
74
75
76
def to_pooling_params(self):
    return PoolingParams(
        truncate_prompt_tokens=self.truncate_prompt_tokens,
        use_activation=get_use_activation(self),
    )

ClassificationCompletionRequest

Bases: PoolingBasicRequestMixin, CompletionRequestMixin

Source code in vllm/entrypoints/pooling/classify/protocol.py 
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
class ClassificationCompletionRequest(PoolingBasicRequestMixin, CompletionRequestMixin):
    # --8<-- [start:classification-extra-params]
    softmax: bool | None = Field(
        default=None,
        description="softmax will be deprecated, please use use_activation instead.",
    )

    activation: bool | None = Field(
        default=None,
        description="activation will be deprecated, please use use_activation instead.",
    )

    use_activation: bool | None = Field(
        default=None,
        description="Whether to use activation for classification outputs. "
        "Default is True.",
    )
    # --8<-- [end:classification-extra-params]

    def to_pooling_params(self):
        return PoolingParams(
            truncate_prompt_tokens=self.truncate_prompt_tokens,
            use_activation=get_use_activation(self),
        )

activation class-attribute instance-attribute 

activation: bool | None = Field(
    default=None,
    description="activation will be deprecated, please use use_activation instead.",
)

softmax class-attribute instance-attribute 

softmax: bool | None = Field(
    default=None,
    description="softmax will be deprecated, please use use_activation instead.",
)

use_activation class-attribute instance-attribute 

use_activation: bool | None = Field(
    default=None,
    description="Whether to use activation for classification outputs. Default is True.",
)

to_pooling_params 

to_pooling_params()
Source code in vllm/entrypoints/pooling/classify/protocol.py 
41
42
43
44
45
def to_pooling_params(self):
    return PoolingParams(
        truncate_prompt_tokens=self.truncate_prompt_tokens,
        use_activation=get_use_activation(self),
    )

ClassificationData

Bases: OpenAIBaseModel

Source code in vllm/entrypoints/pooling/classify/protocol.py 
84
85
86
87
88
class ClassificationData(OpenAIBaseModel):
    index: int
    label: str | None
    probs: list[float]
    num_classes: int

index instance-attribute 

index: int

label instance-attribute 

label: str | None

num_classes instance-attribute 

num_classes: int

probs instance-attribute 

probs: list[float]

ClassificationResponse

Bases: OpenAIBaseModel

Source code in vllm/entrypoints/pooling/classify/protocol.py 
91
92
93
94
95
96
97
class ClassificationResponse(OpenAIBaseModel):
    id: str = Field(default_factory=lambda: f"classify-{random_uuid()}")
    object: str = "list"
    created: int = Field(default_factory=lambda: int(time.time()))
    model: str
    data: list[ClassificationData]
    usage: UsageInfo

created class-attribute instance-attribute 

created: int = Field(default_factory=lambda: int(time()))

data instance-attribute 

data: list[ClassificationData]

id class-attribute instance-attribute 

id: str = Field(
    default_factory=lambda: f"classify-{random_uuid()}"
)

model instance-attribute 

model: str

object class-attribute instance-attribute 

object: str = 'list'

usage instance-attribute 

usage: UsageInfo