Hub Python 函式庫文件
推論類型
並獲得增強的文件體驗
開始使用
推論類型
本頁列出了 Hugging Face Hub 支援的每項任務可用的類型（例如 dataclasses）。每項任務皆使用 JSON 結構描述（JSON schema）進行定義，這些類型則是根據這些結構描述所產生——部分類型會因 Python 的需求而有所客製。請造訪 @huggingface.js/tasks 以查找各任務的 JSON 結構描述。
函式庫的這個部分仍在開發中，未來版本將會持續改進。
audio_classification
class huggingface_hub.AudioClassificationInput
< 原始碼 >( inputs: str parameters: huggingface_hub.inference._generated.types.audio_classification.AudioClassificationParameters | None = None )
音訊分類推論的輸入
音訊分類推論的輸出
class huggingface_hub.AudioClassificationParameters
< 原始碼 >( function_to_apply: typing.Optional[ForwardRef('AudioClassificationOutputTransform')] = None top_k: int | None = None )
音訊分類的其他推論參數
audio_to_audio
音訊到音訊推論的輸入
class huggingface_hub.AudioToAudioOutputElement
< 原始碼 >( blob: typing.Any content_type: str label: str )
音訊到音訊任務的推論輸出。一個包含其標籤的生成音訊檔案。
automatic_speech_recognition
class huggingface_hub.AutomaticSpeechRecognitionGenerationParameters
< 原始碼 >( do_sample: bool | None = None early_stopping: typing.Union[bool, ForwardRef('AutomaticSpeechRecognitionEarlyStoppingEnum'), NoneType] = None epsilon_cutoff: float | None = None eta_cutoff: float | None = None max_length: int | None = None max_new_tokens: int | None = None min_length: int | None = None min_new_tokens: int | None = None num_beam_groups: int | None = None num_beams: int | None = None penalty_alpha: float | None = None temperature: float | None = None top_k: int | None = None top_p: float | None = None typical_p: float | None = None use_cache: bool | None = None )
文字生成過程的參數化
class huggingface_hub.AutomaticSpeechRecognitionInput
< 原始碼 >( inputs: str parameters: huggingface_hub.inference._generated.types.automatic_speech_recognition.AutomaticSpeechRecognitionParameters | None = None )
自動語音辨識推論的輸入
class huggingface_hub.AutomaticSpeechRecognitionOutput
< 原始碼 >( text: str chunks: list[huggingface_hub.inference._generated.types.automatic_speech_recognition.AutomaticSpeechRecognitionOutputChunk] | None = None )
自動語音辨識任務的推論輸出
class huggingface_hub.AutomaticSpeechRecognitionParameters
< 原始碼 >( generation_parameters: huggingface_hub.inference._generated.types.automatic_speech_recognition.AutomaticSpeechRecognitionParameters | None = None return_timestamps: bool | None = None )
自動語音辨識的其他推論參數
chat_completion
class huggingface_hub.ChatCompletionInput
< 原始碼 >( messages: list frequency_penalty: float | None = None logit_bias: list[float] | None = None logprobs: bool | None = None max_tokens: int | None = None model: str | None = None n: int | None = None presence_penalty: float | None = 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: int | None = None stop: list[str] | None = None stream: bool | None = None stream_options: huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputStreamOptions | None = None temperature: float | None = None tool_choice: typing.Union[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputToolChoiceClass, ForwardRef('ChatCompletionInputToolChoiceEnum'), NoneType] = None tool_prompt: str | None = None tools: list[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputTool] | None = None top_logprobs: int | None = None top_p: float | None = None )
聊天完成（Chat Completion）輸入。根據 TGI 規格自動產生。欲知詳情，請參閱 https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts。
class huggingface_hub.ChatCompletionInputFunctionDefinition
< 原始碼 >( name: str parameters: typing.Any description: str | None = None )
class huggingface_hub.ChatCompletionInputJSONSchema
< 原始碼 >( name: str description: str | None = None schema: dict[str, object] | None = None strict: bool | None = None )
class huggingface_hub.ChatCompletionInputMessage
< 原始碼 >( role: str content: list[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputMessageChunk] | str | None = None name: str | None = None tool_calls: list[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputToolCall] | None = None )
class huggingface_hub.ChatCompletionInputMessageChunk
< 原始碼 >( type: ChatCompletionInputMessageChunkType image_url: huggingface_hub.inference._generated.types.chat_completion.ChatCompletionInputURL | None = None text: str | None = None )
class huggingface_hub.ChatCompletionInputResponseFormatJSONObject
< 原始碼 >( type: typing.Literal['json_object'] )
class huggingface_hub.ChatCompletionInputResponseFormatJSONSchema
< 原始碼 >( type: typing.Literal['json_schema'] json_schema: ChatCompletionInputJSONSchema )
class huggingface_hub.ChatCompletionInputResponseFormatText
< 原始碼 >( type: typing.Literal['text'] )
class huggingface_hub.ChatCompletionInputStreamOptions
< 原始碼 >( include_usage: bool | None = None )
class huggingface_hub.ChatCompletionInputTool
< 原始碼 >( function: ChatCompletionInputFunctionDefinition type: str )
class huggingface_hub.ChatCompletionInputToolCall
< 原始碼 >( function: ChatCompletionInputFunctionDefinition id: str type: str )
class huggingface_hub.ChatCompletionInputToolChoiceClass
< 原始碼 >( function: ChatCompletionInputFunctionName )
class huggingface_hub.ChatCompletionOutput
< 原始碼 >( choices: list created: int id: str model: str system_fingerprint: str usage: ChatCompletionOutputUsage )
聊天完成輸出。根據 TGI 規格自動產生。欲知詳情，請參閱 https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts。
class huggingface_hub.ChatCompletionOutputComplete
< 原始碼 >( finish_reason: str index: int message: ChatCompletionOutputMessage logprobs: huggingface_hub.inference._generated.types.chat_completion.ChatCompletionOutputLogprobs | None = None )
class huggingface_hub.ChatCompletionOutputFunctionDefinition
< 原始碼 >( arguments: str name: str description: str | None = None )
class huggingface_hub.ChatCompletionOutputLogprob
< 原始碼 >( logprob: float token: str top_logprobs: list )
class huggingface_hub.ChatCompletionOutputMessage
< 原始碼 >( role: str content: str | None = None reasoning: str | None = None tool_call_id: str | None = None tool_calls: list[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionOutputToolCall] | None = None )
class huggingface_hub.ChatCompletionOutputToolCall
< 原始碼 >( function: ChatCompletionOutputFunctionDefinition id: str type: str )
class huggingface_hub.ChatCompletionOutputUsage
< 原始碼 >( completion_tokens: int prompt_tokens: int total_tokens: int )
class huggingface_hub.ChatCompletionStreamOutput
< 原始碼 >( choices: list created: int id: str model: str system_fingerprint: str usage: huggingface_hub.inference._generated.types.chat_completion.ChatCompletionStreamOutputUsage | None = None )
聊天完成串流輸出。根據 TGI 規格自動產生。欲知詳情，請參閱 https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts。
class huggingface_hub.ChatCompletionStreamOutputChoice
< 原始碼 >( delta: ChatCompletionStreamOutputDelta index: int finish_reason: str | None = None logprobs: huggingface_hub.inference._generated.types.chat_completion.ChatCompletionStreamOutputLogprobs | None = None )
class huggingface_hub.ChatCompletionStreamOutputDelta
< 原始碼 >( role: str content: str | None = None reasoning: str | None = None tool_call_id: str | None = None tool_calls: list[huggingface_hub.inference._generated.types.chat_completion.ChatCompletionStreamOutputDeltaToolCall] | None = None )
class huggingface_hub.ChatCompletionStreamOutputDeltaToolCall
< 來源 >( function: ChatCompletionStreamOutputFunction id: str index: int type: str )
class huggingface_hub.ChatCompletionStreamOutputFunction
< 來源 >( arguments: str name: str | None = None )
class huggingface_hub.ChatCompletionStreamOutputLogprob
< 來源 >( logprob: float token: str top_logprobs: list )
class huggingface_hub.ChatCompletionStreamOutputUsage
< 來源 >( completion_tokens: int prompt_tokens: int total_tokens: int )
depth_estimation
class huggingface_hub.DepthEstimationInput
< 來源 >( inputs: typing.Any parameters: dict[str, typing.Any] | None = None )
深度估計推理的輸入
class huggingface_hub.DepthEstimationOutput
< 來源 >( depth: typing.Any predicted_depth: typing.Any )
深度估計任務的推理輸出
document_question_answering
class huggingface_hub.DocumentQuestionAnsweringInput
< 來源 >( inputs: DocumentQuestionAnsweringInputData parameters: huggingface_hub.inference._generated.types.document_question_answering.DocumentQuestionAnsweringParameters | None = None )
文件問答推理的輸入
一對用於回答的 (文件, 問題)
class huggingface_hub.DocumentQuestionAnsweringOutputElement
< 來源 >( answer: str end: int score: float start: int )
文件問答任務的推理輸出
class huggingface_hub.DocumentQuestionAnsweringParameters
< 來源 >( doc_stride: int | None = None handle_impossible_answer: bool | None = None lang: str | None = None max_answer_len: int | None = None max_question_len: int | None = None max_seq_len: int | None = None top_k: int | None = None word_boxes: list[list[float] | str] | None = None )
文件問答的額外推理參數
feature_extraction
class huggingface_hub.FeatureExtractionInput
< 來源 >( inputs: list[str] | str normalize: bool | None = None prompt_name: str | None = None truncate: bool | None = None truncation_direction: typing.Optional[ForwardRef('FeatureExtractionInputTruncationDirection')] = None )
特徵提取輸入。由 TEI 規範自動生成。欲了解更多詳情，請查看 https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tei-import.ts。
fill_mask
class huggingface_hub.FillMaskInput
< 來源 >( inputs: str parameters: huggingface_hub.inference._generated.types.fill_mask.FillMaskParameters | None = None )
填充遮罩 (Fill Mask) 推理的輸入
class huggingface_hub.FillMaskOutputElement
< 來源 >( score: float sequence: str token: int token_str: typing.Any fill_mask_output_token_str: str | None = None )
填充遮罩任務的推理輸出
class huggingface_hub.FillMaskParameters
< 來源 >( targets: list[str] | None = None top_k: int | None = None )
填充遮罩的額外推理參數
image_classification
class huggingface_hub.ImageClassificationInput
< 來源 >( inputs: str parameters: huggingface_hub.inference._generated.types.image_classification.ImageClassificationParameters | None = None )
圖像分類推理的輸入
圖像分類任務的推理輸出
class huggingface_hub.ImageClassificationParameters
< 來源 >( function_to_apply: typing.Optional[ForwardRef('ImageClassificationOutputTransform')] = None top_k: int | None = None )
圖像分類的額外推理參數
image_segmentation
class huggingface_hub.ImageSegmentationInput
< 來源 >( inputs: str parameters: huggingface_hub.inference._generated.types.image_segmentation.ImageSegmentationParameters | None = None )
圖像分割推理的輸入
class huggingface_hub.ImageSegmentationOutputElement
< 來源 >( label: str mask: str score: float | None = None )
圖像分割任務的推理輸出，包含預測的遮罩/片段
class huggingface_hub.ImageSegmentationParameters
< 來源 >( mask_threshold: float | None = None overlap_mask_area_threshold: float | None = None subtask: typing.Optional[ForwardRef('ImageSegmentationSubtask')] = None threshold: float | None = None )
圖像分割的額外推理參數
image_text_to_image
class huggingface_hub.ImageTextToImageInput
< 來源 >( inputs: str | None = None parameters: huggingface_hub.inference._generated.types.image_text_to_image.ImageTextToImageParameters | None = None )
圖像到圖像 (Image Text To Image) 推理的輸入。必須提供輸入 (圖像) 或提示詞 (在參數中)，或者兩者皆提供。
圖像到圖像任務的推理輸出
class huggingface_hub.ImageTextToImageParameters
< 來源 >( guidance_scale: float | None = None negative_prompt: str | None = None num_inference_steps: int | None = None prompt: str | None = None seed: int | None = None target_size: huggingface_hub.inference._generated.types.image_text_to_image.ImageTextToImageTargetSize | None = None )
圖像到圖像的額外推理參數
輸出圖像的像素大小。此參數僅受部分提供者和特定模型支援。若不支援，將會被忽略。
image_text_to_video
class huggingface_hub.ImageTextToVideoInput
< 來源 >( inputs: str | None = None parameters: huggingface_hub.inference._generated.types.image_text_to_video.ImageTextToVideoParameters | None = None )
圖像到影片 (Image Text To Video) 推理的輸入。必須提供輸入 (圖像) 或提示詞 (在參數中)，或者兩者皆提供。
圖像到影片任務的推理輸出
class huggingface_hub.ImageTextToVideoParameters
< 來源 >( guidance_scale: float | None = None negative_prompt: str | None = None num_frames: float | None = None num_inference_steps: int | None = None prompt: str | None = None seed: int | None = None target_size: huggingface_hub.inference._generated.types.image_text_to_video.ImageTextToVideoTargetSize | None = None )
圖像到影片的額外推理參數
輸出影片影格的像素大小。
image_to_image
class huggingface_hub.ImageToImageInput
< 來源 >( inputs: str parameters: huggingface_hub.inference._generated.types.image_to_image.ImageToImageParameters | None = None )
圖像到圖像推理的輸入
圖像到圖像任務的推理輸出
class huggingface_hub.ImageToImageParameters
< 來源 >( guidance_scale: float | None = None negative_prompt: str | None = None num_inference_steps: int | None = None prompt: str | None = None target_size: huggingface_hub.inference._generated.types.image_to_image.ImageToImageTargetSize | None = None )
圖像到圖像的額外推理參數
輸出圖像的像素大小。此參數僅受部分提供者和特定模型支援。若不支援，將會被忽略。
image_to_text
class huggingface_hub.ImageToTextGenerationParameters
< 來源 >( do_sample: bool | None = None early_stopping: typing.Union[bool, ForwardRef('ImageToTextEarlyStoppingEnum'), NoneType] = None epsilon_cutoff: float | None = None eta_cutoff: float | None = None max_length: int | None = None max_new_tokens: int | None = None min_length: int | None = None min_new_tokens: int | None = None num_beam_groups: int | None = None num_beams: int | None = None penalty_alpha: float | None = None temperature: float | None = None top_k: int | None = None top_p: float | None = None typical_p: float | None = None use_cache: bool | None = None )
文字生成過程的參數化
class huggingface_hub.ImageToTextInput
< 原始碼 >( inputs: typing.Any parameters: huggingface_hub.inference._generated.types.image_to_text.ImageToTextParameters | None = None )
影像轉文字 (Image To Text) 推論的輸入
class huggingface_hub.ImageToTextOutput
< 原始碼 >( generated_text: typing.Any image_to_text_output_generated_text: str | None = None )
影像轉文字任務的推論輸出
class huggingface_hub.ImageToTextParameters
< 原始碼 >( generation_parameters: huggingface_hub.inference._generated.types.image_to_text.ImageToTextGenerationParameters | None = None max_new_tokens: int | None = None )
影像轉文字的額外推論參數
image_to_video
class huggingface_hub.ImageToVideoInput
< 原始碼 >( inputs: str parameters: huggingface_hub.inference._generated.types.image_to_video.ImageToVideoParameters | None = None )
影像轉影片 (Image To Video) 推論的輸入
影像轉影片任務的推論輸出
class huggingface_hub.ImageToVideoParameters
< 原始碼 >( guidance_scale: float | None = None negative_prompt: str | None = None num_frames: float | None = None num_inference_steps: int | None = None prompt: str | None = None seed: int | None = None target_size: huggingface_hub.inference._generated.types.image_to_video.ImageToVideoTargetSize | None = None )
影像轉影片的額外推論參數
輸出影片影格的像素大小。
object_detection
class huggingface_hub.ObjectDetectionBoundingBox
< 原始碼 >( xmax: int xmin: int ymax: int ymin: int )
預測的邊界框。座標相對於輸入影像的左上角。
class huggingface_hub.ObjectDetectionInput
< 原始碼 >( inputs: str parameters: huggingface_hub.inference._generated.types.object_detection.ObjectDetectionParameters | None = None )
物件偵測 (Object Detection) 推論的輸入
class huggingface_hub.ObjectDetectionOutputElement
< 原始碼 >( box: ObjectDetectionBoundingBox label: str score: float )
物件偵測任務的推論輸出
物件偵測的額外推論參數
question_answering
class huggingface_hub.QuestionAnsweringInput
< 原始碼 >( inputs: QuestionAnsweringInputData parameters: huggingface_hub.inference._generated.types.question_answering.QuestionAnsweringParameters | None = None )
問答 (Question Answering) 推論的輸入
一組 (上下文, 問題) 待回答對
class huggingface_hub.QuestionAnsweringOutputElement
< 原始碼 >( answer: str end: int score: float start: int )
問答任務的推論輸出
class huggingface_hub.QuestionAnsweringParameters
< 原始碼 >( align_to_words: bool | None = None doc_stride: int | None = None handle_impossible_answer: bool | None = None max_answer_len: int | None = None max_question_len: int | None = None max_seq_len: int | None = None top_k: int | None = None )
問答的額外推論參數
sentence_similarity
class huggingface_hub.SentenceSimilarityInput
< 原始碼 >( inputs: SentenceSimilarityInputData parameters: dict[str, typing.Any] | None = None )
句對相似度 (Sentence similarity) 推論的輸入
summarization
class huggingface_hub.SummarizationInput
< 原始碼 >( inputs: str parameters: huggingface_hub.inference._generated.types.summarization.SummarizationParameters | None = None )
摘要生成 (Summarization) 推論的輸入
摘要生成任務的推論輸出
class huggingface_hub.SummarizationParameters
< 原始碼 >( clean_up_tokenization_spaces: bool | None = None generate_parameters: dict[str, typing.Any] | None = None truncation: typing.Optional[ForwardRef('SummarizationTruncationStrategy')] = None )
摘要生成的額外推論參數。
table_question_answering
class huggingface_hub.TableQuestionAnsweringInput
< 原始碼 >( inputs: TableQuestionAnsweringInputData parameters: huggingface_hub.inference._generated.types.table_question_answering.TableQuestionAnsweringParameters | None = None )
表格問答 (Table Question Answering) 推論的輸入
一組 (表格, 問題) 待回答對
class huggingface_hub.TableQuestionAnsweringOutputElement
< 原始碼 >( answer: str cells: list coordinates: list aggregator: str | None = None )
表格問答任務的推論輸出
class huggingface_hub.TableQuestionAnsweringParameters
< 原始碼 >( padding: typing.Optional[ForwardRef('Padding')] = None sequential: bool | None = None truncation: bool | None = None )
表格問答的額外推論參數
text2text_generation
class huggingface_hub.Text2TextGenerationInput
< 原始碼 >( inputs: str parameters: huggingface_hub.inference._generated.types.text2text_generation.Text2TextGenerationParameters | None = None )
文字轉文字 (Text2text Generation) 推論的輸入
class huggingface_hub.Text2TextGenerationOutput
< 原始碼 >( generated_text: typing.Any text2_text_generation_output_generated_text: str | None = None )
文字轉文字任務的推論輸出
class huggingface_hub.Text2TextGenerationParameters
< 原始碼 >( clean_up_tokenization_spaces: bool | None = None generate_parameters: dict[str, typing.Any] | None = None truncation: typing.Optional[ForwardRef('Text2TextGenerationTruncationStrategy')] = None )
文字轉文字生成的額外推論參數
text_classification
class huggingface_hub.TextClassificationInput
< 原始碼 >( inputs: str parameters: huggingface_hub.inference._generated.types.text_classification.TextClassificationParameters | None = None )
文字分類 (Text Classification) 推論的輸入
文字分類任務的推論輸出
class huggingface_hub.TextClassificationParameters
< 原始碼 >( function_to_apply: typing.Optional[ForwardRef('TextClassificationOutputTransform')] = None top_k: int | None = None )
文字分類的額外推論參數
text_generation
class huggingface_hub.TextGenerationInput
< 原始碼 >( inputs: str parameters: huggingface_hub.inference._generated.types.text_generation.TextGenerationInputGenerateParameters | None = None stream: bool | None = None )
文字生成輸入。根據 TGI 規格自動產生。如需更多詳細資訊，請查看 https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts。
class huggingface_hub.TextGenerationInputGenerateParameters
< 原始碼 >( adapter_id: str | None = None best_of: int | None = None decoder_input_details: bool | None = None details: bool | None = None do_sample: bool | None = None frequency_penalty: float | None = None grammar: huggingface_hub.inference._generated.types.text_generation.TextGenerationInputGrammarType | None = None max_new_tokens: int | None = None repetition_penalty: float | None = None return_full_text: bool | None = None seed: int | None = None stop: list[str] | None = None temperature: float | None = None top_k: int | None = None top_n_tokens: int | None = None top_p: float | None = None truncate: int | None = None typical_p: float | None = None watermark: bool | None = None )
class huggingface_hub.TextGenerationOutput
< 原始碼 >( generated_text: str details: huggingface_hub.inference._generated.types.text_generation.TextGenerationOutputDetails | None = None )
文字生成輸出。由 TGI 規範自動產生。欲了解更多詳情，請參閱 https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts。
class huggingface_hub.TextGenerationOutputBestOfSequence
< 原始碼 >( finish_reason: TextGenerationOutputFinishReason generated_text: str generated_tokens: int prefill: list tokens: list seed: int | None = None top_tokens: list[list[huggingface_hub.inference._generated.types.text_generation.TextGenerationOutputToken]] | None = None )
class huggingface_hub.TextGenerationOutputDetails
< 原始碼 >( finish_reason: TextGenerationOutputFinishReason generated_tokens: int prefill: list tokens: list best_of_sequences: list[huggingface_hub.inference._generated.types.text_generation.TextGenerationOutputBestOfSequence] | None = None seed: int | None = None top_tokens: list[list[huggingface_hub.inference._generated.types.text_generation.TextGenerationOutputToken]] | None = None )
class huggingface_hub.TextGenerationOutputToken
< 原始碼 >( id: int logprob: float special: bool text: str )
class huggingface_hub.TextGenerationStreamOutput
< 原始碼 >( index: int token: TextGenerationStreamOutputToken details: huggingface_hub.inference._generated.types.text_generation.TextGenerationStreamOutputStreamDetails | None = None generated_text: str | None = None top_tokens: list[huggingface_hub.inference._generated.types.text_generation.TextGenerationStreamOutputToken] | None = None )
文字生成串流輸出。由 TGI 規範自動產生。欲了解更多詳情，請參閱 https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts。
class huggingface_hub.TextGenerationStreamOutputStreamDetails
< 原始碼 >( finish_reason: TextGenerationOutputFinishReason generated_tokens: int input_length: int seed: int | None = None )
class huggingface_hub.TextGenerationStreamOutputToken
< 原始碼 >( id: int logprob: float special: bool text: str )
text_to_audio
class huggingface_hub.TextToAudioGenerationParameters
< 原始碼 >( do_sample: bool | None = None early_stopping: typing.Union[bool, ForwardRef('TextToAudioEarlyStoppingEnum'), NoneType] = None epsilon_cutoff: float | None = None eta_cutoff: float | None = None max_length: int | None = None max_new_tokens: int | None = None min_length: int | None = None min_new_tokens: int | None = None num_beam_groups: int | None = None num_beams: int | None = None penalty_alpha: float | None = None temperature: float | None = None top_k: int | None = None top_p: float | None = None typical_p: float | None = None use_cache: bool | None = None )
文字生成過程的參數化
class huggingface_hub.TextToAudioInput
< 原始碼 >( inputs: str parameters: huggingface_hub.inference._generated.types.text_to_audio.TextToAudioParameters | None = None )
文字轉語音推論的輸入
文字轉語音任務的推論輸出
class huggingface_hub.TextToAudioParameters
< 原始碼 >( generation_parameters: huggingface_hub.inference._generated.types.text_to_audio.TextToAudioParameters | None = None )
文字轉語音的額外推論參數
text_to_image
class huggingface_hub.TextToImageInput
< 原始碼 >( inputs: str parameters: huggingface_hub.inference._generated.types.text_to_image.TextToImageParameters | None = None )
文字轉圖像推論的輸入
文字轉圖像任務的推論輸出
class huggingface_hub.TextToImageParameters
< 原始碼 >( guidance_scale: float | None = None height: int | None = None negative_prompt: str | None = None num_inference_steps: int | None = None scheduler: str | None = None seed: int | None = None width: int | None = None )
文字轉圖像的額外推論參數
text_to_speech
class huggingface_hub.TextToSpeechGenerationParameters
< 原始碼 >( do_sample: bool | None = None early_stopping: typing.Union[bool, ForwardRef('TextToSpeechEarlyStoppingEnum'), NoneType] = None epsilon_cutoff: float | None = None eta_cutoff: float | None = None max_length: int | None = None max_new_tokens: int | None = None min_length: int | None = None min_new_tokens: int | None = None num_beam_groups: int | None = None num_beams: int | None = None penalty_alpha: float | None = None temperature: float | None = None top_k: int | None = None top_p: float | None = None typical_p: float | None = None use_cache: bool | None = None )
文字生成過程的參數化
class huggingface_hub.TextToSpeechInput
< 原始碼 >( inputs: str parameters: huggingface_hub.inference._generated.types.text_to_speech.TextToSpeechParameters | None = None )
文字轉語音推論的輸入
class huggingface_hub.TextToSpeechOutput
< 原始碼 >( audio: typing.Any sampling_rate: float | None = None )
文字轉語音任務的推論輸出
class huggingface_hub.TextToSpeechParameters
< 原始碼 >( generation_parameters: huggingface_hub.inference._generated.types.text_to_speech.TextToSpeechGenerationParameters | None = None )
文字轉語音的額外推論參數
text_to_video
class huggingface_hub.TextToVideoInput
< 原始碼 >( inputs: str parameters: huggingface_hub.inference._generated.types.text_to_video.TextToVideoParameters | None = None )
文字轉影片推論的輸入
文字轉影片任務的推論輸出
class huggingface_hub.TextToVideoParameters
< 原始碼 >( guidance_scale: float | None = None negative_prompt: list[str] | None = None num_frames: float | None = None num_inference_steps: int | None = None seed: int | None = None )
文字轉影片的額外推論參數
token_classification
class huggingface_hub.TokenClassificationInput
< 原始碼 >( inputs: str parameters: huggingface_hub.inference._generated.types.token_classification.TokenClassificationParameters | None = None )
Token 分類推論的輸入
class huggingface_hub.TokenClassificationOutputElement
< 原始碼 >( end: int score: float start: int word: str entity: str | None = None entity_group: str | None = None )
Token 分類任務的推論輸出
class huggingface_hub.TokenClassificationParameters
< 原始碼 >( aggregation_strategy: typing.Optional[ForwardRef('TokenClassificationAggregationStrategy')] = None ignore_labels: list[str] | None = None stride: int | None = None )
Token 分類的額外推論參數
translation
class huggingface_hub.TranslationInput
< 原始碼 >( inputs: str parameters: huggingface_hub.inference._generated.types.translation.TranslationParameters | None = None )
翻譯推論的輸入
翻譯任務的推論輸出
class huggingface_hub.TranslationParameters
< 原始碼 >( clean_up_tokenization_spaces: bool | None = None generate_parameters: dict[str, typing.Any] | None = None src_lang: str | None = None tgt_lang: str | None = None truncation: typing.Optional[ForwardRef('TranslationTruncationStrategy')] = None )
翻譯的額外推論參數
video_classification
class huggingface_hub.VideoClassificationInput
< 原始碼 >( inputs: typing.Any parameters: huggingface_hub.inference._generated.types.video_classification.VideoClassificationParameters | None = None )
影片分類推論的輸入
影片分類任務的推論輸出
class huggingface_hub.VideoClassificationParameters
< 原始碼 >( frame_sampling_rate: int | None = None function_to_apply: typing.Optional[ForwardRef('VideoClassificationOutputTransform')] = None num_frames: int | None = None top_k: int | None = None )
影片分類的額外推論參數
visual_question_answering
class huggingface_hub.VisualQuestionAnsweringInput
< 原始碼 >( inputs: VisualQuestionAnsweringInputData parameters: huggingface_hub.inference._generated.types.visual_question_answering.VisualQuestionAnsweringParameters | None = None )
視覺問答 (VQA) 推論的輸入資料
一個用於回答的（圖像，問題）對
class huggingface_hub.VisualQuestionAnsweringOutputElement
< 原始碼 >( score: float answer: str | None = None )
視覺問答任務的推論輸出
視覺問答的額外推論參數
zero_shot_classification
class huggingface_hub.ZeroShotClassificationInput
< 原始碼 >( inputs: str parameters: ZeroShotClassificationParameters )
零樣本分類 (Zero-Shot Classification) 推論的輸入資料
零樣本分類任務的推論輸出
class huggingface_hub.ZeroShotClassificationParameters
< 原始碼 >( candidate_labels: list hypothesis_template: str | None = None multi_label: bool | None = None )
零樣本分類的額外推論參數
zero_shot_image_classification
class huggingface_hub.ZeroShotImageClassificationInput
< 原始碼 >( inputs: str parameters: ZeroShotImageClassificationParameters )
零樣本圖像分類 (Zero-Shot Image Classification) 推論的輸入資料
零樣本圖像分類任務的推論輸出
class huggingface_hub.ZeroShotImageClassificationParameters
< 原始碼 >( candidate_labels: list hypothesis_template: str | None = None )
零樣本圖像分類的額外推論參數
zero_shot_object_detection
class huggingface_hub.ZeroShotObjectDetectionBoundingBox
< 原始碼 >( xmax: int xmin: int ymax: int ymin: int )
預測的邊界框。座標相對於輸入影像的左上角。
class huggingface_hub.ZeroShotObjectDetectionInput
< 原始碼 >( inputs: str parameters: ZeroShotObjectDetectionParameters )
零樣本物件偵測 (Zero-Shot Object Detection) 推論的輸入資料
class huggingface_hub.ZeroShotObjectDetectionOutputElement
< 原始碼 >( box: ZeroShotObjectDetectionBoundingBox label: str score: float )
零樣本物件偵測任務的推論輸出
零樣本物件偵測的額外推論參數