# transformers_zh_docs **Repository Path**: bukcn/transformers_zh_docs ## Basic Information - **Project Name**: transformers_zh_docs - **Description**: No description available - **Primary Language**: Unknown - **License**: Not specified - **Default Branch**: master - **Homepage**: None - **GVP Project**: No ## Statistics - **Stars**: 0 - **Forks**: 0 - **Created**: 2025-08-27 - **Last Updated**: 2025-08-27 ## Categories & Tags **Categories**: Uncategorized **Tags**: None ## README 本项目为🤗HuggingFace transformers 库的中文文档，仅仅针对英文文档进行了翻译工作，版权归HuggingFace团队所有。 项目会不定期优化翻译质量和排版，欢迎大家捉虫和贡献，也欢迎issues和star. 根据官网的英文目录，对应的中文目录如下： **1. 开始使用** - [🤗Transformers简介](docs_zh/index.md) - [快速开始](docs_zh/quicktour.md) - [安装](docs_zh/installation.md) **2. 教程** - [通过pipline进行推理](docs_zh/pipeline_tutorial.md) - [使用AutoClass编写可移植代码](docs_zh/autoclass_tutorial.md) - [数据预处理](docs_zh/preprocessing.md) - [调优预训练模型](docs_zh/training.md) - [通过脚本训练](docs_zh/run_scripts.md) - [通过🤗Accelerate设置分布式训练](docs_zh/accelerate.md) - [通过🤗PEFT加载和训练adapters](docs_zh/peft.md) - [分享你的模型](docs_zh/model_sharing.md) **3. 任务指南** 3.1 自然语言处理 - [文本分类](docs_zh/tasks/sequence_classification.md) - [Token分类](docs_zh/tasks/token_classification.md) - [问答](docs_zh/tasks/question_answering.md) - [语言模型](docs_zh/tasks/language_modeling.md) - [mask语言模型](docs_zh/tasks/masked_language_modeling.md) - [机器翻译](docs_zh/tasks/translation.md) - [文本摘要](docs_zh/tasks/summarization.md) - [多项选择](docs_zh/tasks/multiple_choice.md) 3.2 语音处理 - [语音分类](docs_zh/tasks/audio_classification.md) - [语音识别](docs_zh/tasks/asr.md) 3.3 机器视觉 - [图像分类](docs_zh/tasks/image_classification.md) - [图像语义分割](docs_zh/tasks/semantic_segmentation.md) - [视频分类](docs_zh/tasks/video_classification.md) - [目标检测](docs_zh/tasks/object_detection.md) - [零样本目标检测](docs_zh/tasks/zero_shot_object_detection.md) - [零样本图像分类](docs_zh/tasks/zero_shot_image_classification.md) - [深度估计](docs_zh/tasks/monocular_depth_estimation.md) 3.4 多模态 - [图像描述](docs_zh/tasks/image_captioning.md) - [文档阅读理解](docs_zh/tasks/document_question_answering.md) - [图像问答](docs_zh/tasks/visual_question_answering.md) - [文本转语音tts](docs_zh/tasks/text-to-speech.md) 3.5 生成 - [自定义生成策略](docs_zh/generation_strategies.md) 3.6 提示 - [通过IDEFICS处理图像任务](docs_zh/tasks/idefics.md) **4. 开发者指南：** - [通过🤗Tokenizers实现快速分词](docs_zh/fast_tokenizers.md) - [多语言模型推理](docs_zh/multilingual.md) - [使用模型特定的API](docs_zh/create_a_model.md) - [分享自定义的模型](docs_zh/custom_models.md) - [chat模型模板](docs_zh/chat_templating.md) - [通过Amazon SageMaker训练模型](docs_zh/serialization.md) - [导出到ONNX](docs_zh/serialization.md) - [导出到TFLite](docs_zh/tflite.md) - [导出到TorchScript](docs_zh/torchscript.md) - [基准测试](docs_zh/benchmarks.md) - [笔记示例](docs_zh/notebooks.md) - [社区资源](docs_zh/community.md) - [自定义Tools和Prompts](docs_zh/custom_tools.md) - [问题排查](docs_zh/troubleshooting.md) **5. 性能和可拓展性：** [5.1 概述](docs_zh/Overview.md) 5.2 高效训练技巧 - [单个GPU上高效训练的方法和工具](docs_zh/perf_train_gpu_one.md) - [多个GPU和并行计算](docs_zh/perf_train_gpu_many.md) - [CPU上高效训练](docs_zh/perf_train_cpu.md) - [分布式CPU训练](docs_zh/perf_train_cpu_many.md) - [TPU上训练](docs_zh/perf_train_tpu.md) - [使用TensorFlow在TPU上训练](docs_zh/perf_train_tpu_tf.md) - [专用硬件训练](docs_zh/perf_train_special.md) - [用于训练的自定义硬件](docs_zh/perf_hardware.md) - [使用Trainer API进行超参数搜索](docs_zh/hpo_train.md) 5.3 优化推理 - [在CPU上进行推理](docs_zh/perf_infer_cpu.md) - [在单个GPU上进行推理](docs_zh/perf_infer_gpu_one.md) - [在多个GPU上进行推理](docs_zh/perf_infer_gpu_many.md) - [在专用硬件上进行推理](docs_zh/perf_infer_special.md) 5.4 其他内容 - [实例化一个大型模型](docs_zh/big_models.md) - [故障排查](docs_zh/debugging.md) - [TensorFlow模型的XLA集成](docs_zh/tf_xla.md) - [使用`torch.compile](docs_zh/.md)`优化推理](docs_zh/perf_torch_compile.md) **6. 给transformers贡献** - [如何贡献给🤗Transformers](docs_zh/contributing.md) - [如何向🤗Transformers添加模型](docs_zh/add_new_model.md) - [如何将🤗Transformers模型转换为TensorFlow](docs_zh/add_tensorflow_model.md) - [如何向🤗Transformers添加pipline](docs_zh/add_new_pipeline.md) - [测试](docs_zh/testing.md) - [pull request 的检查](docs_zh/pr_checks.md) **7. 概念指南** - [哲学](docs_zh/philosophy.md) - [术语表](docs_zh/glossary.md) - [🤗Transformers能做什么](docs_zh/task_summary.md) - [🤗Transformers如何解决任务](docs_zh/tasks_explained.md) - [Transformer模型系列](docs_zh/model_summary.md) - [分词器概述](docs_zh/tokenizer_summary.md) - [注意力机制](docs_zh/attention.md) - [填充和截断](docs_zh/pad_truncation.md) - [BERTology](docs_zh/bertology.md) - [固定长度模型的困惑度](docs_zh/perplexity.md) - [用于Web服务器推论的流水线](docs_zh/pipeline_webserver.md) - [模型训练解剖学](docs_zh/model_memory_anatomy.md) **7 API** **7.1 主要的类** - [Agents and Tools](docs_zh/main_classes/agent.md) - [Auto Classes](docs_zh/model_doc/auto.md) - [Callbacks](docs_zh/main_classes/callback.md) - [Configuration](docs_zh/main_classes/configuration.md) - [Data Collator](docs_zh/main_classes/data_collator.md) - [Keras callbacks](docs_zh/main_classes/keras_callbacks.md) - [Logging](docs_zh/main_classes/logging.md) - [Models](docs_zh/main_classes/model.md) - [Text Generation](docs_zh/main_classes/text_generation.md) - [ONNX](docs_zh/main_classes/onnx.md) - [Optimization](docs_zh/main_classes/optimizer_schedules.md) - [Model outputs](docs_zh/main_classes/output.md) - [Pipelines](docs_zh/main_classes/pipelines.md) - [Processors](docs_zh/main_classes/processors.md) - [Quantization](docs_zh/main_classes/quantization.md) - [Tokenizer](docs_zh/main_classes/tokenizer.md) - [Trainer](docs_zh/main_classes/trainer.md) - [DeepSpeed Integration](docs_zh/main_classes/deepspeed.md) - [Feature Extractor](docs_zh/main_classes/feature_extractor.md) - [Image Processor](docs_zh/main_classes/image_processor.md) **7.2 模型** **7.2.1 文本模型** - [ALBERT](docs_zh/model_doc/albert.md) - [BART](docs_zh/model_doc/bart.md) - [BARThez](docs_zh/model_doc/barthez.md) - [BARTpho](docs_zh/model_doc/bartpho.md) - [BERT](docs_zh/model_doc/bert.md) - [BertGeneration](docs_zh/model_doc/bert-generation.md) - [BertJapanese](docs_zh/model_doc/bert-japanese.md) - [Bertweet](docs_zh/model_doc/bertweet.md) - [BigBird](docs_zh/model_doc/big_bird.md) - [BigBirdPegasus](docs_zh/model_doc/bigbird_pegasus.md) - [BioGpt](docs_zh/model_doc/biogpt.md) - [Blenderbot](docs_zh/model_doc/blenderbot.md) - [Blenderbot Small](docs_zh/model_doc/blenderbot-small.md) - [BLOOM](docs_zh/model_doc/bloom.md) - [BORT](docs_zh/model_doc/bort.md) - [ByT5](docs_zh/model_doc/byt5.md) - [CamemBERT](docs_zh/model_doc/camembert.md) - [CANINE](docs_zh/model_doc/canine.md) - [CodeGen](docs_zh/model_doc/codegen.md) - [CodeLlama](docs_zh/model_doc/code_llama.md) - [ConvBERT](docs_zh/model_doc/convbert.md) - [CPM](docs_zh/model_doc/cpm.md) - [CPMANT](docs_zh/model_doc/cpmant.md) - [CTRL](docs_zh/model_doc/ctrl.md) - [DeBERTa](docs_zh/model_doc/deberta.md) - [DeBERTa-v2](docs_zh/model_doc/deberta-v2.md) - [DialoGPT](docs_zh/model_doc/dialogpt.md) - [DistilBERT](docs_zh/model_doc/distilbert.md) - [DPR](docs_zh/model_doc/dpr.md) - [ELECTRA](docs_zh/model_doc/electra.md) - [Encoder Decoder Models](docs_zh/model_doc/encoder-decoder.md) - [ERNIE](docs_zh/model_doc/ernie.md) - [ErnieM](docs_zh/model_doc/ernie_m.md) - [ESM](docs_zh/model_doc/esm.md) - [Falcon](docs_zh/model_doc/falcon.md) - [FLAN-T5](docs_zh/model_doc/flan-t5.md) - [FLAN-UL2](docs_zh/model_doc/flan-ul2.md) - [FlauBERT](docs_zh/model_doc/flaubert.md) - [FNet](docs_zh/model_doc/fnet.md) - [FSMT](docs_zh/model_doc/fsmt.md) - [Funnel Transformer](docs_zh/model_doc/funnel.md) - [GPT](docs_zh/model_doc/openai-gpt.md) - [GPT Neo](docs_zh/model_doc/gpt_neo.md) - [GPT NeoX](docs_zh/model_doc/gpt_neox.md) - [GPT NeoX Japanese](docs_zh/model_doc/gpt_neox_japanese.md) - [GPT-J](docs_zh/model_doc/gptj.md) - [GPT2](docs_zh/model_doc/gpt2.md) - [GPTBigCode](docs_zh/model_doc/gpt_bigcode.md) - [GPTSAN Japanese](docs_zh/model_doc/gptsan-japanese.md) - [GPTSw3](docs_zh/model_doc/gpt-sw3.md) - [HerBERT](docs_zh/model_doc/herbert.md) - [I-BERT](docs_zh/model_doc/ibert.md) - [Jukebox](docs_zh/model_doc/jukebox.md) - [LED](docs_zh/model_doc/led.md) - [LLaMA](docs_zh/model_doc/llama.md) - [Llama2](docs_zh/model_doc/llama2.md) - [Longformer](docs_zh/model_doc/longformer.md) - [LongT5](docs_zh/model_doc/longt5.md) - [LUKE](docs_zh/model_doc/luke.md) - [M2M100](docs_zh/model_doc/m2m_100.md) - [MarianMT](docs_zh/model_doc/marian.md) - [MarkupLM](docs_zh/model_doc/markuplm.md) - [MBart and MBart-50](docs_zh/model_doc/mbart.md) - [MEGA](docs_zh/model_doc/mega.md) - [MegatronBERT](docs_zh/model_doc/megatron-bert.md) - [MegatronGPT2](docs_zh/model_doc/megatron_gpt2.md) - [mLUKE](docs_zh/model_doc/mluke.md) - [MobileBERT](docs_zh/model_doc/mobilebert.md) - [MPNet](docs_zh/model_doc/mpnet.md) - [MPT](docs_zh/model_doc/mpt.md) - [MRA](docs_zh/model_doc/mra.md) - [MT5](docs_zh/model_doc/mt5.md) - [MVP](docs_zh/model_doc/mvp.md) - [NEZHA](docs_zh/model_doc/nezha.md) - [NLLB](docs_zh/model_doc/nllb.md) - [NLLB-MoE](docs_zh/model_doc/nllb-moe.md) - [Nyströmformer](docs_zh/model_doc/nystromformer.md) - [Open-Llama](docs_zh/model_doc/open-llama.md) - [OPT](docs_zh/model_doc/opt.md) - [Pegasus](docs_zh/model_doc/pegasus.md) - [PEGASUS-X](docs_zh/model_doc/pegasus_x.md) - [Persimmon](docs_zh/model_doc/persimmon.md) - [PhoBERT](docs_zh/model_doc/phobert.md) - [PLBart](docs_zh/model_doc/plbart.md) - [ProphetNet](docs_zh/model_doc/prophetnet.md) - [QDQBert](docs_zh/model_doc/qdqbert.md) - [RAG](docs_zh/model_doc/rag.md) - [REALM](docs_zh/model_doc/realm.md) - [Reformer](docs_zh/model_doc/reformer.md) - [RemBERT](docs_zh/model_doc/rembert.md) - [RetriBERT](docs_zh/model_doc/retribert.md) - [RoBERTa](docs_zh/model_doc/roberta.md) - [RoBERTa-PreLayerNorm](docs_zh/model_doc/roberta-prelayernorm.md) - [RoCBert](docs_zh/model_doc/roc_bert.md) - [RoFormer](docs_zh/model_doc/roformer.md) - [RWKV](docs_zh/model_doc/rwkv.md) - [Splinter](docs_zh/model_doc/splinter.md) - [SqueezeBERT](docs_zh/model_doc/squeezebert.md) - [SwitchTransformers](docs_zh/model_doc/switch_transformers.md) - [T5](docs_zh/model_doc/t5.md) - [T5v1.1](docs_zh/model_doc/t5v1.1.md) - [TAPEX](docs_zh/model_doc/tapex.md) - [Transformer XL](docs_zh/model_doc/transfo-xl.md) - [UL2](docs_zh/model_doc/ul2.md) - [UMT5](docs_zh/model_doc/umt5.md) - [X-MOD](docs_zh/model_doc/xmod.md) - [XGLM](docs_zh/model_doc/xglm.md) - [XLM](docs_zh/model_doc/xlm.md) - [XLM-ProphetNet](docs_zh/model_doc/xlm-prophetnet.md) - [XLM-RoBERTa](docs_zh/model_doc/xlm-roberta.md) - [XLM-RoBERTa-XL](docs_zh/model_doc/xlm-roberta-xl.md) - [XLM-V](docs_zh/model_doc/xlm-v.md) - [XLNet](docs_zh/model_doc/xlnet.md) - [YOSO](docs_zh/model_doc/yoso.md) **7.2.2 视觉模型** - [BEiT](docs_zh/model_doc/beit.md) - [BiT](docs_zh/model_doc/bit.md) - [Conditional DETR](docs_zh/model_doc/conditional_detr.md) - [ConvNeXT](docs_zh/model_doc/convnext.md) - [ConvNeXTV2](docs_zh/model_doc/convnextv2.md) - [CvT](docs_zh/model_doc/cvt.md) - [Deformable DETR](docs_zh/model_doc/deformable_detr.md) - [DeiT](docs_zh/model_doc/deit.md) - [DETA](docs_zh/model_doc/deta.md) - [DETR](docs_zh/model_doc/detr.md) - [DiNAT](docs_zh/model_doc/dinat.md) - [DINO V2](docs_zh/model_doc/dinov2.md) - [DiT](docs_zh/model_doc/dit.md) - [DPT](docs_zh/model_doc/dpt.md) - [EfficientFormer](docs_zh/model_doc/efficientformer.md) - [EfficientNet](docs_zh/model_doc/efficientnet.md) - [FocalNet](docs_zh/model_doc/focalnet.md) - [GLPN](docs_zh/model_doc/glpn.md) - [ImageGPT](docs_zh/model_doc/imagegpt.md) - [LeViT](docs_zh/model_doc/levit.md) - [Mask2Former](docs_zh/model_doc/mask2former.md) - [MaskFormer](docs_zh/model_doc/maskformer.md) - [MobileNetV1](docs_zh/model_doc/mobilenet_v1.md) - [MobileNetV2](docs_zh/model_doc/mobilenet_v2.md) - [MobileViT](docs_zh/model_doc/mobilevit.md) - [MobileViTV2](docs_zh/model_doc/mobilevitv2.md) - [NAT](docs_zh/model_doc/nat.md) - [PoolFormer](docs_zh/model_doc/poolformer.md) - [Pyramid Vision Transformer](docs_zh/model_doc/pvt.md) - [RegNet](docs_zh/model_doc/regnet.md) - [ResNet](docs_zh/model_doc/resnet.md) - [SegFormer](docs_zh/model_doc/segformer.md) - [SwiftFormer](docs_zh/model_doc/swiftformer.md) - [Swin Transformer](docs_zh/model_doc/swin.md) - [Swin Transformer V2](docs_zh/model_doc/swinv2.md) - [Swin2SR](docs_zh/model_doc/swin2sr.md) - [Table Transformer](docs_zh/model_doc/table-transformer.md) - [TimeSformer](docs_zh/model_doc/timesformer.md) - [UperNet](docs_zh/model_doc/upernet.md) - [VAN](docs_zh/model_doc/van.md) - [VideoMAE](docs_zh/model_doc/videomae.md) - [Vision Transformer](docs_zh/model_doc/vit.md) - [ViT Hybrid](docs_zh/model_doc/vit_hybrid.md) - [ViTDet](docs_zh/model_doc/vitdet.md) - [ViTMAE](docs_zh/model_doc/vit_mae.md) - [ViTMatte](docs_zh/model_doc/vitmatte.md) - [ViTMSN](docs_zh/model_doc/vit_msn.md) - [ViViT](docs_zh/model_doc/vivit.md) - [YOLOS](docs_zh/model_doc/yolos.md) **7.2.3 语音模型** - [Audio Spectrogram Transformer](docs_zh/model_doc/audio-spectrogram-transformer.md) - [Bark](docs_zh/model_doc/bark.md) - [CLAP](docs_zh/model_doc/clap.md) - [EnCodec](docs_zh/model_doc/encodec.md) - [Hubert](docs_zh/model_doc/hubert.md) - [MCTCT](docs_zh/model_doc/mctct.md) - [MMS](docs_zh/model_doc/mms.md) - [MusicGen](docs_zh/model_doc/musicgen.md) - [Pop2Piano](docs_zh/model_doc/pop2piano.md) - [SEW](docs_zh/model_doc/sew.md) - [SEW-D](docs_zh/model_doc/sew-d.md) - [Speech2Text](docs_zh/model_doc/speech_to_text.md) - [Speech2Text2](docs_zh/model_doc/speech_to_text_2.md) - [SpeechT5](docs_zh/model_doc/speecht5.md) - [UniSpeech](docs_zh/model_doc/unispeech.md) - [UniSpeech-SAT](docs_zh/model_doc/unispeech-sat.md) - [VITS](docs_zh/model_doc/vits.md) - [Wav2Vec2](docs_zh/model_doc/wav2vec2.md) - [Wav2Vec2-Conformer](docs_zh/model_doc/wav2vec2-conformer.md) - [Wav2Vec2Phoneme](docs_zh/model_doc/wav2vec2_phoneme.md) - [WavLM](docs_zh/model_doc/wavlm.md) - [Whisper](docs_zh/model_doc/whisper.md) - [XLS-R](docs_zh/model_doc/xls_r.md) - [XLSR-Wav2Vec2](docs_zh/model_doc/xlsr_wav2vec2.md) **7.2.4 多模态模型** - [ALIGN](docs_zh/model_doc/align.md) - [AltCLIP](docs_zh/model_doc/altclip.md) - [BLIP](docs_zh/model_doc/blip.md) - [BLIP-2](docs_zh/model_doc/blip-2.md) - [BridgeTower](docs_zh/model_doc/bridgetower.md) - [BROS](docs_zh/model_doc/bros.md) - [Chinese-CLIP](docs_zh/model_doc/chinese_clip.md) - [CLIP](docs_zh/model_doc/clip.md) - [CLIPSeg](docs_zh/model_doc/clipseg.md) - [Data2Vec](docs_zh/model_doc/data2vec.md) - [DePlot](docs_zh/model_doc/deplot.md) - [Donut](docs_zh/model_doc/donut.md) - [FLAVA](docs_zh/model_doc/flava.md) - [GIT](docs_zh/model_doc/git.md) - [GroupViT](docs_zh/model_doc/groupvit.md) - [IDEFICS](docs_zh/model_doc/idefics.md) - [InstructBLIP](docs_zh/model_doc/instructblip.md) - [LayoutLM](docs_zh/model_doc/layoutlm.md) - [LayoutLMV2](docs_zh/model_doc/layoutlmv2.md) - [LayoutLMV3](docs_zh/model_doc/layoutlmv3.md) - [LayoutXLM](docs_zh/model_doc/layoutxlm.md) - [LiLT](docs_zh/model_doc/lilt.md) - [LXMERT](docs_zh/model_doc/lxmert.md) - [MatCha](docs_zh/model_doc/matcha.md) - [MGP-STR](docs_zh/model_doc/mgp-str.md) - [OneFormer](docs_zh/model_doc/oneformer.md) - [OWL-ViT](docs_zh/model_doc/owlvit.md) - [Perceiver](docs_zh/model_doc/perceiver.md) - [Pix2Struct](docs_zh/model_doc/pix2struct.md) - [Segment Anything](docs_zh/model_doc/sam.md) - [Speech Encoder Decoder Models](docs_zh/model_doc/speech-encoder-decoder.md) - [TAPAS](docs_zh/model_doc/tapas.md) - [TrOCR](docs_zh/model_doc/trocr.md) - [TVLT](docs_zh/model_doc/tvlt.md) - [ViLT](docs_zh/model_doc/vilt.md) - [Vision Encoder Decoder Models](docs_zh/model_doc/vision-encoder-decoder.md) - [Vision Text Dual Encoder](docs_zh/model_doc/vision-text-dual-encoder.md) - [VisualBERT](docs_zh/model_doc/visual_bert.md) - [X-CLIP](docs_zh/model_doc/xclip.md) **7.2.5 强化学习模型** - [Decision Transformer](docs_zh/model_doc/decision_transformer.md) - [Trajectory Transformer](docs_zh/model_doc/trajectory_transformer.md) **7.2.6 时序模型** - [Autoformer](docs_zh/model_doc/autoformer.md) - [Informer](docs_zh/model_doc/informer.md) - [Time Series Transformer](docs_zh/model_doc/time_series_transformer.md) **7.2.6 图模型** - [Graphormer](docs_zh/model_doc/graphormer.md) **7.3 内部工具** - [Custom Layers and Utilities](docs_zh/internal/modeling_utils.md) - [Utilities for pipelines](docs_zh/internal/pipelines_utils.md) - [Utilities for Tokenizers](docs_zh/internal/tokenization_utils.md) - [Utilities for Trainer](docs_zh/internal/trainer_utils.md) - [Utilities for Generation](docs_zh/internal/generation_utils.md) - [Utilities for Image Processors](docs_zh/internal/image_processing_utils.md) - [Utilities for Audio processing](docs_zh/internal/audio_utils.md) - [General Utilities](docs_zh/internal/file_utils.md) - [Utilities for Time Series](docs_zh/internal/time_series_utils.md)