本文属于机器翻译版本。若本译文内容与英语原文存在差异，则一律以英文原文为准。

# 特定于任务的模型
<a name="jumpstart-models"></a>

JumpStart 支持十五种最流行的问题类型的任务特定模型。在支持的问题类型中，共有十三种类型与视觉和 NLP 相关。有八种问题类型支持增量训练和微调。有关增量训练和超参数调整的更多信息，请参阅 [SageMaker AI 自动模型](https://docs.amazonaws.cn/sagemaker/latest/dg/automatic-model-tuning.html)调整。 JumpStart 还支持四种常用的表格数据建模算法。

您可以从 Studio 或 Studio Classic 的 JumpStart 登录页面搜索和浏览模特。当您选择模型时，模型详细信息页面会提供有关该模型的信息，您可以通过几个步骤来训练和部署模型。描述部分介绍了您可以通过模型完成的任务、预期的输入和输出类型以及微调模型所需的数据类型。

您还可以通过 [SageMaker Python 软件开发工具包](https://sagemaker.readthedocs.io/en/stable/overview.html#use-prebuilt-models-with-sagemaker-jumpstart)以编程方式使用模型。有关所有可用型号的列表，请参阅[JumpStart可用型号表](https://sagemaker.readthedocs.io/en/v2.132.0/doc_utils/pretrainedmodels.html)。

下表汇总了问题类型列表及其示例 Jupyter 笔记本的链接。


| 问题类型  | 支持使用预训练模型进行推理  | 可在自定义数据集上训练  | 支持的框架  | 示例笔记本  | 
| --- | --- | --- | --- | --- | 
| 图像分类  | 支持  | 是  | PyTorch, TensorFlow | [简介 JumpStart -图像分类](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_image_classification/Amazon_JumpStart_Image_Classification.ipynb) | 
| 对象检测  | 支持  | 是  | PyTorch, TensorFlow, MXNet | [简介 JumpStart -物体检测](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_object_detection/Amazon_JumpStart_Object_Detection.ipynb) | 
| 语义分割  | 支持  | 是  | MXNet  | [简介 JumpStart -语义分割](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_semantic_segmentation/Amazon_JumpStart_Semantic_Segmentation.ipynb) | 
| 实例分段  | 支持  | 是  | MXNet  | [简介 JumpStart -实例分割](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_instance_segmentation/Amazon_JumpStart_Instance_Segmentation.ipynb) | 
| 图像嵌入  | 是  | 否  | TensorFlow, MXNet | [简介 JumpStart -图像嵌入](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_image_embedding/Amazon_JumpStart_Image_Embedding.ipynb) | 
| 文本分类  | 支持  | 是  | TensorFlow | [简介 JumpStart -文本分类](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_text_classification/Amazon_JumpStart_Text_Classification.ipynb) | 
| 句子对分类  | 支持  | 是  | TensorFlow，Hugging Face | [简介 JumpStart -句子对分类](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_sentence_pair_classification/Amazon_JumpStart_Sentence_Pair_Classification.ipynb) | 
| 问题回答  | 支持  | 是  | PyTorch，Hugging Face | [简介 JumpStart — 问题解答](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_question_answering/Amazon_JumpStart_Question_Answering.ipynb) | 
| 指定实体识别  | 是  | 否  | Hugging Face  | [简介 JumpStart -命名实体识别](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_named_entity_recognition/Amazon_JumpStart_Named_Entity_Recognition.ipynb) | 
| 文本摘要  | 是  | 否  | Hugging Face  | [简介 JumpStart -文本摘要](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_text_summarization/Amazon_JumpStart_Text_Summarization.ipynb) | 
| 文本生成  | 是  | 否  | Hugging Face  | [简介 JumpStart -文本生成](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_text_generation/Amazon_JumpStart_Text_Generation.ipynb) | 
| 机器翻译  | 是  | 否  | Hugging Face  | [简介 JumpStart -机器翻译](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_machine_translation/Amazon_JumpStart_Machine_Translation.ipynb) | 
| 文本嵌入  | 是  | 否  | TensorFlow, MXNet | [简介 JumpStart -文本嵌入](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_text_embedding/Amazon_JumpStart_Text_Embedding.ipynb) | 
| 表格分类  | 支持  | 是  | LightGBM、、 AutoGluon-Tabular CatBoost、 XGBoost、Linear Learner TabTransformer | [简介 JumpStart ——表格分类——LightGBM， CatBoost](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/lightgbm_catboost_tabular/Amazon_Tabular_Classification_LightGBM_CatBoost.ipynb)<br />[简介 JumpStart -表格分类- XGBoost，线性学习器](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/xgboost_linear_learner_tabular/Amazon_Tabular_Classification_XGBoost_LinearLearner.ipynb)<br />[简介 JumpStart -表格分类-学员 AutoGluon](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/autogluon_tabular/Amazon_Tabular_Classification_AutoGluon.ipynb)<br />[简介 JumpStart -表格分类-学员 TabTransformer](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/tabtransformer_tabular/Amazon_Tabular_Classification_TabTransformer.ipynb) | 
| 表格回归  | 支持  | 是  | LightGBM、、 AutoGluon-Tabular CatBoost、 XGBoost、Linear Learner TabTransformer | [简介 JumpStart ——表格回归——lightGBM， CatBoost](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/lightgbm_catboost_tabular/Amazon_Tabular_Regression_LightGBM_CatBoost.ipynb)<br />[简介 JumpStart — 表格回归- XGBoost，线性学习器](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/xgboost_linear_learner_tabular/Amazon_Tabular_Regression_XGBoost_LinearLearner.ipynb)<br />[简介 JumpStart — 表格回归-学员 AutoGluon ](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/autogluon_tabular/Amazon_Tabular_Regression_AutoGluon.ipynb)<br />[简介 JumpStart — 表格回归-学员 TabTransformer](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/tabtransformer_tabular/Amazon_Tabular_Regression_TabTransformer.ipynb) |