Hub Python 库文档
推理类型
并获得增强的文档体验
开始使用
推理类型
本页面列出了 Hugging Face Hub 支持的每项任务的可用类型（例如数据类）。每项任务都使用 JSON Schema 指定，这些类型是根据这些 Schema 生成的，并根据 Python 要求进行了一些定制。访问 @huggingface.js/tasks 以查找每项任务的 JSON Schema。
该库的这一部分仍在开发中，将在未来版本中得到改进。
audio_classification
class huggingface_hub.AudioClassificationInput
< source >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.audio_classification.AudioClassificationParameters] = None )
音频分类推理的输入
音频分类推理的输出
class huggingface_hub.AudioClassificationParameters
< source >( function_to_apply: typing.Optional[ForwardRef('AudioClassificationOutputTransform')] = None top_k: typing.Optional[int] = None )
音频分类的额外推理参数
audio_to_audio
音频到音频推理的输入
class huggingface_hub.AudioToAudioOutputElement
< source >( blob: typing.Any content_type: str label: str )
音频到音频任务的推理输出 一个生成的带标签的音频文件。
自动语音识别
class huggingface_hub.AutomaticSpeechRecognitionGenerationParameters
< source >( do_sample: typing.Optional[bool] = None early_stopping: typing.Union[bool, ForwardRef('AutomaticSpeechRecognitionEarlyStoppingEnum'), NoneType] = None epsilon_cutoff: typing.Optional[float] = None eta_cutoff: typing.Optional[float] = None max_length: typing.Optional[int] = None max_new_tokens: typing.Optional[int] = None min_length: typing.Optional[int] = None min_new_tokens: typing.Optional[int] = None num_beam_groups: typing.Optional[int] = None num_beams: typing.Optional[int] = None penalty_alpha: typing.Optional[float] = None temperature: typing.Optional[float] = None top_k: typing.Optional[int] = None top_p: typing.Optional[float] = None typical_p: typing.Optional[float] = None use_cache: typing.Optional[bool] = None )
文本生成过程的参数化
class huggingface_hub.AutomaticSpeechRecognitionInput
< source >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.automatic_speech_recognition.AutomaticSpeechRecognitionParameters] = None )
自动语音识别推理的输入
class huggingface_hub.AutomaticSpeechRecognitionOutput
< source >( text: str chunks: typing.Optional[typing.List[huggingface_hub.inference._generated.types.automatic_speech_recognition.AutomaticSpeechRecognitionOutputChunk]] = None )
自动语音识别任务的推理输出
class huggingface_hub.AutomaticSpeechRecognitionOutputChunk
< source >( text: str timestamp: typing.List[float] )
class huggingface_hub.AutomaticSpeechRecognitionParameters
< source >( generation_parameters: typing.Optional[huggingface_hub.inference._generated.types.automatic_speech_recognition.AutomaticSpeechRecognitionGenerationParameters] = None return_timestamps: typing.Optional[bool] = None )
自动语音识别的额外推理参数
聊天补全
class huggingface_hub.ChatCompletionInput
< source >( messages: typing.List[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputMessage] frequency_penalty: typing.Optional[float] = None logit_bias: typing.Optional[typing.List[float]] = None logprobs: typing.Optional[bool] = None max_tokens: typing.Optional[int] = None model: typing.Optional[str] = None n: typing.Optional[int] = None presence_penalty: typing.Optional[float] = None response_format: typing.Union[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputResponseFormatText, huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputResponseFormatJSONSchema, huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputResponseFormatJSONObject, NoneType] = None seed: typing.Optional[int] = None stop: typing.Optional[typing.List[str]] = None stream: typing.Optional[bool] = None stream_options: typing.Optional[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputStreamOptions] = None temperature: typing.Optional[float] = None tool_choice: typing.Union[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputToolChoiceClass, ForwardRef('ChatCompletionInputToolChoiceEnum'), NoneType] = None tool_prompt: typing.Optional[str] = None tools: typing.Optional[typing.List[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputTool]] = None top_logprobs: typing.Optional[int] = None top_p: typing.Optional[float] = None )
聊天补全输入。从 TGI 规范自动生成。更多详细信息，请查看 https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts。
class huggingface_hub.ChatCompletionInputFunctionDefinition
< source >( name: str parameters: typing.Any description: typing.Optional[str] = None )
class huggingface_hub.ChatCompletionInputJSONSchema
< source >( name: str description: typing.Optional[str] = None schema: typing.Optional[typing.Dict[str, object]] = None strict: typing.Optional[bool] = None )
class huggingface_hub.ChatCompletionInputMessage
< source >( role: str content: typing.Union[typing.List[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputMessageChunk], str, NoneType] = None name: typing.Optional[str] = None tool_calls: typing.Optional[typing.List[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputToolCall]] = None )
class huggingface_hub.ChatCompletionInputMessageChunk
< source >( type: ChatCompletionInputMessageChunkType image_url: typing.Optional[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputURL] = None text: typing.Optional[str] = None )
class huggingface_hub.ChatCompletionInputResponseFormatJSONObject
< source >( type: typing.Literal['json_object'] )
class huggingface_hub.ChatCompletionInputResponseFormatJSONSchema
< source >( type: typing.Literal['json_schema'] json_schema: ChatCompletionInputJSONSchema )
class huggingface_hub.ChatCompletionInputResponseFormatText
< source >( type: typing.Literal['text'] )
class huggingface_hub.ChatCompletionInputStreamOptions
< source >( include_usage: typing.Optional[bool] = None )
class huggingface_hub.ChatCompletionInputTool
< source >( function: ChatCompletionInputFunctionDefinition type: str )
class huggingface_hub.ChatCompletionInputToolCall
< source >( function: ChatCompletionInputFunctionDefinition id: str type: str )
class huggingface_hub.ChatCompletionInputToolChoiceClass
< source >( function: ChatCompletionInputFunctionName )
class huggingface_hub.ChatCompletionOutput
< source >( choices: typing.List[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionOutputComplete] created: int id: str model: str system_fingerprint: str usage: ChatCompletionOutputUsage )
聊天补全输出。从 TGI 规范自动生成。更多详细信息，请查看 https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts。
class huggingface_hub.ChatCompletionOutputComplete
< source >( finish_reason: str index: int message: ChatCompletionOutputMessage logprobs: typing.Optional[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionOutputLogprobs] = None )
class huggingface_hub.ChatCompletionOutputFunctionDefinition
< source >( arguments: str name: str description: typing.Optional[str] = None )
class huggingface_hub.ChatCompletionOutputLogprob
< source >( logprob: float token: str top_logprobs: typing.List[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionOutputTopLogprob] )
class huggingface_hub.ChatCompletionOutputLogprobs
< 来源 >( content: typing.List[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionOutputLogprob] )
class huggingface_hub.ChatCompletionOutputMessage
< 来源 >( role: str content: typing.Optional[str] = None tool_call_id: typing.Optional[str] = None tool_calls: typing.Optional[typing.List[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionOutputToolCall]] = None )
class huggingface_hub.ChatCompletionOutputToolCall
< 来源 >( function: ChatCompletionOutputFunctionDefinition id: str type: str )
class huggingface_hub.ChatCompletionOutputUsage
< 来源 >( completion_tokens: int prompt_tokens: int total_tokens: int )
class huggingface_hub.ChatCompletionStreamOutput
< 来源 >( choices: typing.List[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionStreamOutputChoice] created: int id: str model: str system_fingerprint: str usage: typing.Optional[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionStreamOutputUsage] = None )
聊天完成流输出。从TGI规范自动生成。有关更多详细信息，请查看https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts。
class huggingface_hub.ChatCompletionStreamOutputChoice
< 来源 >( delta: ChatCompletionStreamOutputDelta index: int finish_reason: typing.Optional[str] = None logprobs: typing.Optional[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionStreamOutputLogprobs] = None )
class huggingface_hub.ChatCompletionStreamOutputDelta
< 来源 >( role: str content: typing.Optional[str] = None tool_call_id: typing.Optional[str] = None tool_calls: typing.Optional[typing.List[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionStreamOutputDeltaToolCall]] = None )
class huggingface_hub.ChatCompletionStreamOutputDeltaToolCall
< 来源 >( function: ChatCompletionStreamOutputFunction id: str index: int type: str )
class huggingface_hub.ChatCompletionStreamOutputFunction
< 来源 >( arguments: str name: typing.Optional[str] = None )
class huggingface_hub.ChatCompletionStreamOutputLogprob
< 来源 >( logprob: float token: str top_logprobs: typing.List[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionStreamOutputTopLogprob] )
class huggingface_hub.ChatCompletionStreamOutputLogprobs
< 来源 >( content: typing.List[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionStreamOutputLogprob] )
class huggingface_hub.ChatCompletionStreamOutputUsage
< 来源 >( completion_tokens: int prompt_tokens: int total_tokens: int )
depth_estimation
class huggingface_hub.DepthEstimationInput
< 来源 >( inputs: typing.Any parameters: typing.Optional[typing.Dict[str, typing.Any]] = None )
深度估计推理的输入
class huggingface_hub.DepthEstimationOutput
< 来源 >( depth: typing.Any predicted_depth: typing.Any )
深度估计任务的推理输出
document_question_answering
class huggingface_hub.DocumentQuestionAnsweringInput
< 来源 >( inputs: DocumentQuestionAnsweringInputData parameters: typing.Optional[huggingface_hub.inference._generated.types.document_question_answering.DocumentQuestionAnsweringParameters] = None )
文档问答推理的输入
一个（文档，问题）对以回答
class huggingface_hub.DocumentQuestionAnsweringOutputElement
< 来源 >( answer: str end: int score: float start: int )
文档问答任务的推理输出
class huggingface_hub.DocumentQuestionAnsweringParameters
< 来源 >( doc_stride: typing.Optional[int] = None handle_impossible_answer: typing.Optional[bool] = None lang: typing.Optional[str] = None max_answer_len: typing.Optional[int] = None max_question_len: typing.Optional[int] = None max_seq_len: typing.Optional[int] = None top_k: typing.Optional[int] = None word_boxes: typing.Optional[typing.List[typing.Union[typing.List[float], str]]] = None )
文档问答的其他推理参数
feature_extraction
class huggingface_hub.FeatureExtractionInput
< 来源 >( inputs: typing.Union[typing.List[str], str] normalize: typing.Optional[bool] = None prompt_name: typing.Optional[str] = None truncate: typing.Optional[bool] = None truncation_direction: typing.Optional[ForwardRef('FeatureExtractionInputTruncationDirection')] = None )
特征提取输入。从TEI规范自动生成。有关更多详细信息，请查看https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tei-import.ts。
fill_mask
class huggingface_hub.FillMaskInput
< 来源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.fill_mask.FillMaskParameters] = None )
填充掩码推理的输入
class huggingface_hub.FillMaskOutputElement
< 来源 >( score: float sequence: str token: int token_str: typing.Any fill_mask_output_token_str: typing.Optional[str] = None )
填充掩码任务的推理输出
class huggingface_hub.FillMaskParameters
< 来源 >( targets: typing.Optional[typing.List[str]] = None top_k: typing.Optional[int] = None )
填充掩码的其他推理参数
image_classification
class huggingface_hub.ImageClassificationInput
< 来源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.image_classification.ImageClassificationParameters] = None )
图像分类推理的输入
图像分类任务的推理输出
class huggingface_hub.ImageClassificationParameters
< 来源 >( function_to_apply: typing.Optional[ForwardRef('ImageClassificationOutputTransform')] = None top_k: typing.Optional[int] = None )
图像分类的其他推理参数
image_segmentation
class huggingface_hub.ImageSegmentationInput
< 来源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.image_segmentation.ImageSegmentationParameters] = None )
图像分割推理的输入
class huggingface_hub.ImageSegmentationOutputElement
< 来源 >( label: str mask: str score: typing.Optional[float] = None )
图像分割任务的推理输出，预测的掩码/片段
class huggingface_hub.ImageSegmentationParameters
< 来源 >( mask_threshold: typing.Optional[float] = None overlap_mask_area_threshold: typing.Optional[float] = None subtask: typing.Optional[ForwardRef('ImageSegmentationSubtask')] = None threshold: typing.Optional[float] = None )
图像分割的其他推理参数
image_to_image
class huggingface_hub.ImageToImageInput
< 来源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.image_to_image.ImageToImageParameters] = None )
图像到图像推理的输入
图像到图像任务的推理输出
class huggingface_hub.ImageToImageParameters
< 来源 >( guidance_scale: typing.Optional[float] = None negative_prompt: typing.Optional[str] = None num_inference_steps: typing.Optional[int] = None prompt: typing.Optional[str] = None target_size: typing.Optional[huggingface_hub.inference._generated.types.image_to_image.ImageToImageTargetSize] = None )
图像到图像的其他推理参数
输出图像的像素大小。
image_to_text
class huggingface_hub.ImageToTextGenerationParameters
< 来源 >( do_sample: typing.Optional[bool] = None early_stopping: typing.Union[bool, ForwardRef('ImageToTextEarlyStoppingEnum'), NoneType] = None epsilon_cutoff: typing.Optional[float] = None eta_cutoff: typing.Optional[float] = None max_length: typing.Optional[int] = None max_new_tokens: typing.Optional[int] = None min_length: typing.Optional[int] = None min_new_tokens: typing.Optional[int] = None num_beam_groups: typing.Optional[int] = None num_beams: typing.Optional[int] = None penalty_alpha: typing.Optional[float] = None temperature: typing.Optional[float] = None top_k: typing.Optional[int] = None top_p: typing.Optional[float] = None typical_p: typing.Optional[float] = None use_cache: typing.Optional[bool] = None )
文本生成过程的参数化
class huggingface_hub.ImageToTextInput
< 来源 >( inputs: typing.Any parameters: typing.Optional[huggingface_hub.inference._generated.types.image_to_text.ImageToTextParameters] = None )
图像到文本推理的输入
class huggingface_hub.ImageToTextOutput
< 源 >( generated_text: typing.Any image_to_text_output_generated_text: typing.Optional[str] = None )
图像转文本任务的推理输出
类 huggingface_hub.ImageToTextParameters
< 源 >( generation_parameters: typing.Optional[huggingface_hub.inference._generated.types.image_to_text.ImageToTextGenerationParameters] = None max_new_tokens: typing.Optional[int] = None )
图像转文本的其他推理参数
image_to_video (图像转视频)
类 huggingface_hub.ImageToVideoInput
< 源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.image_to_video.ImageToVideoParameters] = None )
图像转视频推理的输入
图像转视频任务的推理输出
类 huggingface_hub.ImageToVideoParameters
< 源 >( guidance_scale: typing.Optional[float] = None negative_prompt: typing.Optional[str] = None num_frames: typing.Optional[float] = None num_inference_steps: typing.Optional[int] = None prompt: typing.Optional[str] = None seed: typing.Optional[int] = None target_size: typing.Optional[huggingface_hub.inference._generated.types.image_to_video.ImageToVideoTargetSize] = None )
图像转视频的其他推理参数
输出视频帧的像素大小。
object_detection (目标检测)
预测的边界框。坐标是相对于输入图像左上角的。
类 huggingface_hub.ObjectDetectionInput
< 源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.object_detection.ObjectDetectionParameters] = None )
目标检测推理的输入
类 huggingface_hub.ObjectDetectionOutputElement
< 源 >( box: ObjectDetectionBoundingBox label: str score: float )
目标检测任务的推理输出
目标检测的其他推理参数
question_answering (问答)
类 huggingface_hub.QuestionAnsweringInput
< 源 >( inputs: QuestionAnsweringInputData parameters: typing.Optional[huggingface_hub.inference._generated.types.question_answering.QuestionAnsweringParameters] = None )
问答推理的输入
一个 (上下文, 问题) 对用于回答
类 huggingface_hub.QuestionAnsweringOutputElement
< 源 >( answer: str end: int score: float start: int )
问答任务的推理输出
类 huggingface_hub.QuestionAnsweringParameters
< 源 >( align_to_words: typing.Optional[bool] = None doc_stride: typing.Optional[int] = None handle_impossible_answer: typing.Optional[bool] = None max_answer_len: typing.Optional[int] = None max_question_len: typing.Optional[int] = None max_seq_len: typing.Optional[int] = None top_k: typing.Optional[int] = None )
问答的其他推理参数
sentence_similarity (句子相似度)
类 huggingface_hub.SentenceSimilarityInput
< 源 >( inputs: SentenceSimilarityInputData parameters: typing.Optional[typing.Dict[str, typing.Any]] = None )
句子相似度推理的输入
类 huggingface_hub.SentenceSimilarityInputData
< 源 >( sentences: typing.List[str] source_sentence: str )
summarization (摘要)
类 huggingface_hub.SummarizationInput
< 源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.summarization.SummarizationParameters] = None )
摘要推理的输入
摘要任务的推理输出
类 huggingface_hub.SummarizationParameters
< 源 >( clean_up_tokenization_spaces: typing.Optional[bool] = None generate_parameters: typing.Optional[typing.Dict[str, typing.Any]] = None truncation: typing.Optional[ForwardRef('SummarizationTruncationStrategy')] = None )
摘要的其他推理参数。
table_question_answering (表格问答)
类 huggingface_hub.TableQuestionAnsweringInput
< 源 >( inputs: TableQuestionAnsweringInputData parameters: typing.Optional[huggingface_hub.inference._generated.types.table_question_answering.TableQuestionAnsweringParameters] = None )
表格问答推理的输入
类 huggingface_hub.TableQuestionAnsweringInputData
< 源 >( question: str table: typing.Dict[str, typing.List[str]] )
一个 (表格, 问题) 对用于回答
类 huggingface_hub.TableQuestionAnsweringOutputElement
< 源 >( answer: str cells: typing.List[str] coordinates: typing.List[typing.List[int]] aggregator: typing.Optional[str] = None )
表格问答任务的推理输出
类 huggingface_hub.TableQuestionAnsweringParameters
< 源 >( padding: typing.Optional[ForwardRef('Padding')] = None sequential: typing.Optional[bool] = None truncation: typing.Optional[bool] = None )
表格问答的其他推理参数
text2text_generation (文本到文本生成)
类 huggingface_hub.Text2TextGenerationInput
< 源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.text2text_generation.Text2TextGenerationParameters] = None )
文本到文本生成推理的输入
类 huggingface_hub.Text2TextGenerationOutput
< 源 >( generated_text: typing.Any text2_text_generation_output_generated_text: typing.Optional[str] = None )
文本到文本生成任务的推理输出
类 huggingface_hub.Text2TextGenerationParameters
< 源 >( clean_up_tokenization_spaces: typing.Optional[bool] = None generate_parameters: typing.Optional[typing.Dict[str, typing.Any]] = None truncation: typing.Optional[ForwardRef('Text2TextGenerationTruncationStrategy')] = None )
文本到文本生成的其他推理参数
text_classification (文本分类)
类 huggingface_hub.TextClassificationInput
< 源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.text_classification.TextClassificationParameters] = None )
文本分类推理的输入
文本分类任务的推理输出
类 huggingface_hub.TextClassificationParameters
< 源 >( function_to_apply: typing.Optional[ForwardRef('TextClassificationOutputTransform')] = None top_k: typing.Optional[int] = None )
文本分类的其他推理参数
text_generation (文本生成)
类 huggingface_hub.TextGenerationInput
< 源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.text_generation.TextGenerationInputGenerateParameters] = None stream: typing.Optional[bool] = None )
文本生成输入。从 TGI 规范自动生成。有关更多详细信息，请查看 https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts。
类 huggingface_hub.TextGenerationInputGenerateParameters
< 源 >( adapter_id: typing.Optional[str] = None best_of: typing.Optional[int] = None decoder_input_details: typing.Optional[bool] = None details: typing.Optional[bool] = None do_sample: typing.Optional[bool] = None frequency_penalty: typing.Optional[float] = None grammar: typing.Optional[huggingface_hub.inference._generated.types.text_generation.TextGenerationInputGrammarType] = None max_new_tokens: typing.Optional[int] = None repetition_penalty: typing.Optional[float] = None return_full_text: typing.Optional[bool] = None seed: typing.Optional[int] = None stop: typing.Optional[typing.List[str]] = None temperature: typing.Optional[float] = None top_k: typing.Optional[int] = None top_n_tokens: typing.Optional[int] = None top_p: typing.Optional[float] = None truncate: typing.Optional[int] = None typical_p: typing.Optional[float] = None watermark: typing.Optional[bool] = None )
class huggingface_hub.TextGenerationOutput
< 源 >( generated_text: str details: typing.Optional[huggingface_hub.inference._generated.types.text_generation.TextGenerationOutputDetails] = None )
文本生成输出。根据TGI规范自动生成。更多详细信息，请查看 https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts。
class huggingface_hub.TextGenerationOutputBestOfSequence
< 源 >( finish_reason: TextGenerationOutputFinishReason generated_text: str generated_tokens: int prefill: typing.List[huggingface_hub.inference._generated.types.text_generation.TextGenerationOutputPrefillToken] tokens: typing.List[huggingface_hub.inference._generated.types.text_generation.TextGenerationOutputToken] seed: typing.Optional[int] = None top_tokens: typing.Optional[typing.List[typing.List[huggingface_hub.inference._generated.types.text_generation.TextGenerationOutputToken]]] = None )
class huggingface_hub.TextGenerationOutputDetails
< 源 >( finish_reason: TextGenerationOutputFinishReason generated_tokens: int prefill: typing.List[huggingface_hub.inference._generated.types.text_generation.TextGenerationOutputPrefillToken] tokens: typing.List[huggingface_hub.inference._generated.types.text_generation.TextGenerationOutputToken] best_of_sequences: typing.Optional[typing.List[huggingface_hub.inference._generated.types.text_generation.TextGenerationOutputBestOfSequence]] = None seed: typing.Optional[int] = None top_tokens: typing.Optional[typing.List[typing.List[huggingface_hub.inference._generated.types.text_generation.TextGenerationOutputToken]]] = None )
class huggingface_hub.TextGenerationOutputToken
< 源 >( id: int logprob: float special: bool text: str )
class huggingface_hub.TextGenerationStreamOutput
< 源 >( index: int token: TextGenerationStreamOutputToken details: typing.Optional[huggingface_hub.inference._generated.types.text_generation.TextGenerationStreamOutputStreamDetails] = None generated_text: typing.Optional[str] = None top_tokens: typing.Optional[typing.List[huggingface_hub.inference._generated.types.text_generation.TextGenerationStreamOutputToken]] = None )
文本生成流输出。根据TGI规范自动生成。更多详细信息，请查看 https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts。
class huggingface_hub.TextGenerationStreamOutputStreamDetails
< 源 >( finish_reason: TextGenerationOutputFinishReason generated_tokens: int input_length: int seed: typing.Optional[int] = None )
class huggingface_hub.TextGenerationStreamOutputToken
< 源 >( id: int logprob: float special: bool text: str )
text_to_audio
class huggingface_hub.TextToAudioGenerationParameters
< 源 >( do_sample: typing.Optional[bool] = None early_stopping: typing.Union[bool, ForwardRef('TextToAudioEarlyStoppingEnum'), NoneType] = None epsilon_cutoff: typing.Optional[float] = None eta_cutoff: typing.Optional[float] = None max_length: typing.Optional[int] = None max_new_tokens: typing.Optional[int] = None min_length: typing.Optional[int] = None min_new_tokens: typing.Optional[int] = None num_beam_groups: typing.Optional[int] = None num_beams: typing.Optional[int] = None penalty_alpha: typing.Optional[float] = None temperature: typing.Optional[float] = None top_k: typing.Optional[int] = None top_p: typing.Optional[float] = None typical_p: typing.Optional[float] = None use_cache: typing.Optional[bool] = None )
文本生成过程的参数化
class huggingface_hub.TextToAudioInput
< 源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.text_to_audio.TextToAudioParameters] = None )
文本到音频推理的输入
文本到音频任务的推理输出
class huggingface_hub.TextToAudioParameters
< 源 >( generation_parameters: typing.Optional[huggingface_hub.inference._generated.types.text_to_audio.TextToAudioGenerationParameters] = None )
文本到音频的其他推理参数
text_to_image
class huggingface_hub.TextToImageInput
< 源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.text_to_image.TextToImageParameters] = None )
文本到图像推理的输入
文本到图像任务的推理输出
class huggingface_hub.TextToImageParameters
< 源 >( guidance_scale: typing.Optional[float] = None height: typing.Optional[int] = None negative_prompt: typing.Optional[str] = None num_inference_steps: typing.Optional[int] = None scheduler: typing.Optional[str] = None seed: typing.Optional[int] = None width: typing.Optional[int] = None )
文本到图像的其他推理参数
text_to_speech
class huggingface_hub.TextToSpeechGenerationParameters
< 源 >( do_sample: typing.Optional[bool] = None early_stopping: typing.Union[bool, ForwardRef('TextToSpeechEarlyStoppingEnum'), NoneType] = None epsilon_cutoff: typing.Optional[float] = None eta_cutoff: typing.Optional[float] = None max_length: typing.Optional[int] = None max_new_tokens: typing.Optional[int] = None min_length: typing.Optional[int] = None min_new_tokens: typing.Optional[int] = None num_beam_groups: typing.Optional[int] = None num_beams: typing.Optional[int] = None penalty_alpha: typing.Optional[float] = None temperature: typing.Optional[float] = None top_k: typing.Optional[int] = None top_p: typing.Optional[float] = None typical_p: typing.Optional[float] = None use_cache: typing.Optional[bool] = None )
文本生成过程的参数化
class huggingface_hub.TextToSpeechInput
< 源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.text_to_speech.TextToSpeechParameters] = None )
文本到语音推理的输入
class huggingface_hub.TextToSpeechOutput
< 源 >( audio: typing.Any sampling_rate: typing.Optional[float] = None )
文本到语音任务的推理输出
class huggingface_hub.TextToSpeechParameters
< 源 >( generation_parameters: typing.Optional[huggingface_hub.inference._generated.types.text_to_speech.TextToSpeechGenerationParameters] = None )
文本到语音的其他推理参数
text_to_video
class huggingface_hub.TextToVideoInput
< 源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.text_to_video.TextToVideoParameters] = None )
文本到视频推理的输入
文本到视频任务的推理输出
class huggingface_hub.TextToVideoParameters
< 源 >( guidance_scale: typing.Optional[float] = None negative_prompt: typing.Optional[typing.List[str]] = None num_frames: typing.Optional[float] = None num_inference_steps: typing.Optional[int] = None seed: typing.Optional[int] = None )
文本到视频的其他推理参数
token_classification
class huggingface_hub.TokenClassificationInput
< 源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.token_classification.TokenClassificationParameters] = None )
令牌分类推理的输入
class huggingface_hub.TokenClassificationOutputElement
< 源 >( end: int score: float start: int word: str entity: typing.Optional[str] = None entity_group: typing.Optional[str] = None )
令牌分类任务的推理输出
class huggingface_hub.TokenClassificationParameters
< 源 >( aggregation_strategy: typing.Optional[ForwardRef('TokenClassificationAggregationStrategy')] = None ignore_labels: typing.Optional[typing.List[str]] = None stride: typing.Optional[int] = None )
令牌分类的其他推理参数
translation
class huggingface_hub.TranslationInput
< 源 >( inputs: str parameters: typing.Optional[huggingface_hub.inference._generated.types.translation.TranslationParameters] = None )
翻译推理的输入
翻译任务的推理输出
class huggingface_hub.TranslationParameters
< 源 >( clean_up_tokenization_spaces: typing.Optional[bool] = None generate_parameters: typing.Optional[typing.Dict[str, typing.Any]] = None src_lang: typing.Optional[str] = None tgt_lang: typing.Optional[str] = None truncation: typing.Optional[ForwardRef('TranslationTruncationStrategy')] = None )
翻译的其他推理参数
video_classification
class huggingface_hub.VideoClassificationInput
< 源 >( inputs: typing.Any parameters: typing.Optional[huggingface_hub.inference._generated.types.video_classification.VideoClassificationParameters] = None )
视频分类推理的输入
视频分类任务的推理输出
class huggingface_hub.VideoClassificationParameters
< 源 >( frame_sampling_rate: typing.Optional[int] = None function_to_apply: typing.Optional[ForwardRef('VideoClassificationOutputTransform')] = None num_frames: typing.Optional[int] = None top_k: typing.Optional[int] = None )
视频分类的其他推理参数
视觉问答
class huggingface_hub.VisualQuestionAnsweringInput
< 源 >( inputs: VisualQuestionAnsweringInputData parameters: typing.Optional[huggingface_hub.inference._generated.types.visual_question_answering.VisualQuestionAnsweringParameters] = None )
视觉问答推理的输入
一个 (图像, 问题) 对用于回答
class huggingface_hub.VisualQuestionAnsweringOutputElement
< 源 >( score: float answer: typing.Optional[str] = None )
视觉问答任务的推理输出
class huggingface_hub.VisualQuestionAnsweringParameters
< 源 >( top_k: typing.Optional[int] = None )
视觉问答的其他推理参数
零样本分类
class huggingface_hub.ZeroShotClassificationInput
< 源 >( inputs: str parameters: ZeroShotClassificationParameters )
零样本分类推理的输入
零样本分类任务的推理输出
class huggingface_hub.ZeroShotClassificationParameters
< 源 >( candidate_labels: typing.List[str] hypothesis_template: typing.Optional[str] = None multi_label: typing.Optional[bool] = None )
零样本分类的其他推理参数
零样本图像分类
class huggingface_hub.ZeroShotImageClassificationInput
< 源 >( inputs: str parameters: ZeroShotImageClassificationParameters )
零样本图像分类推理的输入
零样本图像分类任务的推理输出
class huggingface_hub.ZeroShotImageClassificationParameters
< 源 >( candidate_labels: typing.List[str] hypothesis_template: typing.Optional[str] = None )
零样本图像分类的其他推理参数
零样本目标检测
class huggingface_hub.ZeroShotObjectDetectionBoundingBox
< 源 >( xmax: int xmin: int ymax: int ymin: int )
预测的边界框。坐标是相对于输入图像左上角的。
class huggingface_hub.ZeroShotObjectDetectionInput
< 源 >( inputs: str parameters: ZeroShotObjectDetectionParameters )
零样本目标检测推理的输入
class huggingface_hub.ZeroShotObjectDetectionOutputElement
< 源 >( box: ZeroShotObjectDetectionBoundingBox label: str score: float )
零样本目标检测任务的推理输出
class huggingface_hub.ZeroShotObjectDetectionParameters
< 源 >( candidate_labels: typing.List[str] )
零样本目标检测的其他推理参数