本文為英文版的機器翻譯版本，如內容有任何歧義或不一致之處，概以英文版為準。

# 具體工作模型
<a name="jumpstart-models"></a>

JumpStart 支援十五種最常見問題類型的具體工作模型。在支援的問題類型中，共有 13 種視覺和 NTP 相關類型。有八種問題類型支援增量訓練和微調。如需增量訓練和超參數調整的詳細資訊，請參閱 [SageMaker AI 自動模型微調](https://docs.aws.amazon.com/sagemaker/latest/dg/automatic-model-tuning.html)。JumpStart 也支援四種常用的表格式資料建模演算法。

您可以在 Studio 或 Studio Classic 中從 JumpStart 登陸頁面搜尋和瀏覽模型。當您選取模型時，模型詳細資訊頁面會提供模型的相關資訊，您只需數個步驟即可訓練和部署模型。說明部分描述了您可以對模型執行的操作、預期的輸入和輸出類型，以及微調模型所需的資料類型。

您也可以透過程式設計方式搭配 [SageMaker Python SDK](https://sagemaker.readthedocs.io/en/stable/overview.html#use-prebuilt-models-with-sagemaker-jumpstart) 使用模型。如需所有可用模型的清單，請參閱 [JumpStart 可用模型表格](https://sagemaker.readthedocs.io/en/v2.132.0/doc_utils/pretrainedmodels.html)。

下表摘要列出問題類型及其範例 Jupyter 筆記本的連結。


| 問題類型  | 支援預先訓練模型的推論  | 可在自訂資料集上訓練  | 支援的架構  | 範例筆記本  | 
| --- | --- | --- | --- | --- | 
| Image classification  | 是  | 是  | PyTorch、TensorFlow | [JumpStart 簡介 - 影像分類](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_image_classification/Amazon_JumpStart_Image_Classification.ipynb) | 
| 物件偵測  | 是  | 是  | PyTorch、TensorFlow、MXNet | [JumpStart 簡介 - 物件偵測](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_object_detection/Amazon_JumpStart_Object_Detection.ipynb) | 
| 語意分割  | 是  | 是  | MXNet  | [JumpStart 簡介 - 語意分割](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_semantic_segmentation/Amazon_JumpStart_Semantic_Segmentation.ipynb) | 
| 實例分割  | 是  | 是  | MXNet  | [JumpStart 簡介 - 實例分割](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_instance_segmentation/Amazon_JumpStart_Instance_Segmentation.ipynb) | 
| 圖像嵌入  | 是  | 否  | TensorFlow、MXNet | [JumpStart 簡介 - 圖像嵌入](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_image_embedding/Amazon_JumpStart_Image_Embedding.ipynb) | 
| 文字分類  | 是  | 是  | TensorFlow | [JumpStart 簡介 - 文字分類](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_text_classification/Amazon_JumpStart_Text_Classification.ipynb) | 
| 句子對分類  | 是  | 是  | TensorFlow、Hugging Face | [JumpStart 簡介 - 句子對分類](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_sentence_pair_classification/Amazon_JumpStart_Sentence_Pair_Classification.ipynb) | 
| 回答問題  | 是  | 是  | PyTorch、Hugging Face | [JumpStart 簡介 - 回答問題](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_question_answering/Amazon_JumpStart_Question_Answering.ipynb) | 
| 具名實體辨識  | 是  | 否  | Hugging Face  | [JumpStart 簡介 - 具名實體識別](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_named_entity_recognition/Amazon_JumpStart_Named_Entity_Recognition.ipynb) | 
| 文字摘要  | 是  | 否  | Hugging Face  | [JumpStart 簡介 - 文字摘要](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_text_summarization/Amazon_JumpStart_Text_Summarization.ipynb) | 
| 產生文字  | 是  | 否  | Hugging Face  | [JumpStart 簡介 - 文字產生](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_text_generation/Amazon_JumpStart_Text_Generation.ipynb) | 
| 機器翻譯  | 是  | 否  | Hugging Face  | [JumpStart 簡介 - 機器翻譯](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_machine_translation/Amazon_JumpStart_Machine_Translation.ipynb) | 
| 文字嵌入  | 是  | 否  | TensorFlow、MXNet | [JumpStart 簡介 - 文字嵌入](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart_text_embedding/Amazon_JumpStart_Text_Embedding.ipynb) | 
| 表格分類  | 是  | 是  | LightGBM、CatBoost、XGBoost、AutoGluon-Tabular、TabTransformer、線性學習 | [JumpStart 簡介 - 表格分類 - 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 />[JumpStart 簡介 - 表格分類 - XGBoost、線性學習](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/xgboost_linear_learner_tabular/Amazon_Tabular_Classification_XGBoost_LinearLearner.ipynb)<br />[JumpStart 簡介 - 表格分類 - AutoGluon Learner](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/autogluon_tabular/Amazon_Tabular_Classification_AutoGluon.ipynb)<br />[JumpStart 簡介 - 表格分類 - TabTransformer Learner](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/tabtransformer_tabular/Amazon_Tabular_Classification_TabTransformer.ipynb) | 
| 表格迴歸  | 是  | 是  | LightGBM、CatBoost、XGBoost、AutoGluon-Tabular、TabTransformer、線性學習 | [JumpStart 簡介 - 表格迴歸 - 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 />[JumpStart 簡介 - 表格迴歸 - XGBoost、線性學習](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/xgboost_linear_learner_tabular/Amazon_Tabular_Regression_XGBoost_LinearLearner.ipynb)<br />[JumpStart 簡介 - 表格迴歸 - AutoGluon Learner](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/autogluon_tabular/Amazon_Tabular_Regression_AutoGluon.ipynb)<br />[JumpStart 簡介 - 表格迴歸 - TabTransformer Learner](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/tabtransformer_tabular/Amazon_Tabular_Regression_TabTransformer.ipynb) |