通过

OnnxGenAIPromptExecutionSettings Class

OnnxGenAI prompt execution settings.

Initialize the prompt execution settings.

Constructor

OnnxGenAIPromptExecutionSettings(service_id: str | None = None, *, extension_data: dict[str, Any] = None, function_choice_behavior: FunctionChoiceBehavior | None = None, diversity_penalty: Annotated[float | None, Ge(ge=0.0), Le(le=1.0)] = None, do_sample: bool = False, early_stopping: bool = True, length_penalty: Annotated[float | None, Ge(ge=0.0), Le(le=1.0)] = None, max_length: Annotated[int, Gt(gt=0)] = 3072, min_length: Annotated[int | None, Gt(gt=0)] = None, no_repeat_ngram_size: int = 0, num_beams: Annotated[int | None, Gt(gt=0)] = None, num_return_sequences: Annotated[int | None, Gt(gt=0)] = None, past_present_share_buffer: int = True, repetition_penalty: Annotated[float | None, Ge(ge=0.0), Le(le=1.0)] = None, temperature: Annotated[float | None, Ge(ge=0.0), Le(le=2.0)] = None, top_k: Annotated[int | None, Gt(gt=0)] = None, top_p: Annotated[float | None, Ge(ge=0.0), Le(le=1.0)] = None)

Parameters

Name Description
service_id
str

The service ID to use for the request.

Default value: None
kwargs
Required
Any

Additional keyword arguments, these are attempted to parse into the keys of the specific prompt execution settings.

Keyword-Only Parameters

Name Description
extension_data
Required
function_choice_behavior
Required
diversity_penalty
Required
do_sample
Required
early_stopping
Default value: True
length_penalty
Required
max_length
Default value: 3072
min_length
Required
no_repeat_ngram_size
Required
num_beams
Required
num_return_sequences
Required
past_present_share_buffer
Default value: True
repetition_penalty
Required
temperature
Required
top_k
Required
top_p
Required

Attributes

diversity_penalty 

diversity_penalty: Annotated[float | None, FieldInfo(annotation=NoneType, required=True, metadata=[Ge(ge=0.0), Le(le=1.0)])]

do_sample 

do_sample: bool

early_stopping 

early_stopping: bool

length_penalty 

length_penalty: Annotated[float | None, FieldInfo(annotation=NoneType, required=True, metadata=[Ge(ge=0.0), Le(le=1.0)])]

max_length 

max_length: Annotated[int, FieldInfo(annotation=NoneType, required=True, metadata=[Gt(gt=0)])]

min_length 

min_length: Annotated[int | None, FieldInfo(annotation=NoneType, required=True, metadata=[Gt(gt=0)])]

no_repeat_ngram_size 

no_repeat_ngram_size: int

num_beams 

num_beams: Annotated[int | None, FieldInfo(annotation=NoneType, required=True, metadata=[Gt(gt=0)])]

num_return_sequences 

num_return_sequences: Annotated[int | None, FieldInfo(annotation=NoneType, required=True, metadata=[Gt(gt=0)])]

past_present_share_buffer 

past_present_share_buffer: int

repetition_penalty 

repetition_penalty: Annotated[float | None, FieldInfo(annotation=NoneType, required=True, metadata=[Ge(ge=0.0), Le(le=1.0)])]

temperature 

temperature: Annotated[float | None, FieldInfo(annotation=NoneType, required=True, metadata=[Ge(ge=0.0), Le(le=2.0)])]

top_k 

top_k: Annotated[int | None, FieldInfo(annotation=NoneType, required=True, metadata=[Gt(gt=0)])]

top_p 

top_p: Annotated[float | None, FieldInfo(annotation=NoneType, required=True, metadata=[Ge(ge=0.0), Le(le=1.0)])]