AWS sagemaker documentation change
Summary
Reorganized table columns and updated example ordering in algorithm mapping table. Changed 'SageMaker JumpStart' to 'Amazon SageMaker JumpStart' for branding consistency.
Security assessment
The changes involve structural reorganization of a reference table and branding updates. While the table includes security-relevant use cases like IP anomaly detection and content moderation, these are existing examples presented in a new order rather than new security documentation. No specific vulnerabilities or security improvements are introduced.
Diff
diff --git a/sagemaker/latest/dg/algos.md b/sagemaker/latest/dg/algos.md index 4399a5bac..eb8a08ce1 100644 --- a//sagemaker/latest/dg/algos.md +++ b//sagemaker/latest/dg/algos.md @@ -13 +13 @@ Table: Mapping use cases to built-in algorithms -**Example problems and use cases** | **Learning paradigm or domain** | **Problem types** | **Data input format** | **Built-in algorithms** +**Learning paradigm or domain** | **Problem types** | **Example problems and use cases** | **Data input format** | **Built-in algorithms** @@ -15,18 +15,18 @@ Table: Mapping use cases to built-in algorithms -Here a few examples out of the 15 problem types that can be addressed by the pre-trained models and pre-built solution templates provided by SageMaker JumpStart: Question answering: chatbot that outputs an answer for a given question. Text analysis: analyze texts from models specific to an industry domain such as finance. | [Pre-trained models and pre-built solution templates](https://docs.aws.amazon.com/sagemaker/latest/dg/studio-jumpstart.html) | Image Classification Tabular Classification Tabular Regression Text Classification Object Detection Text Embedding Question Answering Sentence Pair Classification Image Embedding Named Entity Recognition Instance Segmentation Text Generation Text Summarization Semantic Segmentation Machine Translation | Image, Text, Tabular | Popular models, including Mobilenet, YOLO, Faster R-CNN, BERT, lightGBM, and CatBoostFor a list of pre-trained models available, see [JumpStart Models](https://docs.aws.amazon.com/sagemaker/latest/dg/studio-jumpstart.html#jumpstart-models). For a list of pre-built solution templates available, see [JumpStart Solutions](https://docs.aws.amazon.com/sagemaker/latest/dg/studio-jumpstart.html#jumpstart-solutions). -Predict if an item belongs to a category: an email spam filter | Supervised learning | Binary/multi-class classification | Tabular | [AutoGluon-Tabular](./autogluon-tabular.html), [CatBoost](./catboost.html), [Factorization Machines Algorithm](./fact-machines.html), [K-Nearest Neighbors (k-NN) Algorithm](./k-nearest-neighbors.html), [LightGBM](./lightgbm.html), [Linear Learner Algorithm](./linear-learner.html), [TabTransformer](./tabtransformer.html), [XGBoost algorithm with Amazon SageMaker AI](./xgboost.html) -Predict a numeric/continuous value: estimate the value of a house | Regression | Tabular | [AutoGluon-Tabular](./autogluon-tabular.html), [CatBoost](./catboost.html), [Factorization Machines Algorithm](./fact-machines.html), [K-Nearest Neighbors (k-NN) Algorithm](./k-nearest-neighbors.html), [LightGBM](./lightgbm.html), [Linear Learner Algorithm](./linear-learner.html), [TabTransformer](./tabtransformer.html), [XGBoost algorithm with Amazon SageMaker AI](./xgboost.html) -Based on historical data for a behavior, predict future behavior: predict sales on a new product based on previous sales data. | Time-series forecasting | Tabular | [Use the SageMaker AI DeepAR forecasting algorithm](./deepar.html) -Improve the data embeddings of the high-dimensional objects: identify duplicate support tickets or find the correct routing based on similarity of text in the tickets | Embeddings: convert high-dimensional objects into low-dimensional space. | Tabular | [Object2Vec Algorithm](./object2vec.html) -Drop those columns from a dataset that have a weak relation with the label/target variable: the color of a car when predicting its mileage. | Unsupervised learning | Feature engineering: dimensionality reduction | Tabular | [Principal Component Analysis (PCA) Algorithm](./pca.html) -Detect abnormal behavior in application: spot when an IoT sensor is sending abnormal readings | Anomaly detection | Tabular | [Random Cut Forest (RCF) Algorithm](./randomcutforest.html) -Protect your application from suspicious users: detect if an IP address accessing a service might be from a bad actor | IP anomaly detection | Tabular | [IP Insights](./ip-insights.html) -Group similar objects/data together: find high-, medium-, and low-spending customers from their transaction histories | Clustering or grouping | Tabular | [K-Means Algorithm](./k-means.html) -Organize a set of documents into topics (not known in advance): tag a document as belonging to a medical category based on the terms used in the document. | Topic modeling | Text | [Latent Dirichlet Allocation (LDA) Algorithm](./lda.html), [Neural Topic Model (NTM) Algorithm](./ntm.html) -Assign pre-defined categories to documents in a corpus: categorize books in a library into academic disciplines | Textual analysis | Text classification | Text | [BlazingText algorithm](./blazingtext.html), [Text Classification - TensorFlow](./text-classification-tensorflow.html) -Convert text from one language to other: Spanish to English | Machine translationalgorithm | Text | [Sequence-to-Sequence Algorithm](./seq-2-seq.html) -Summarize a long text corpus: an abstract for a research paper | Text summarization | Text | [Sequence-to-Sequence Algorithm](./seq-2-seq.html) -Convert audio files to text: transcribe call center conversations for further analysis | Speech-to-text | Text | [Sequence-to-Sequence Algorithm](./seq-2-seq.html) -Label/tag an image based on the content of the image: alerts about adult content in an image | Image processing | Image and multi-label classification | Image | [Image Classification - MXNet](./image-classification.html) -Classify something in an image using transfer learning. | Image classification | Image | [Image Classification - TensorFlow](./image-classification-tensorflow.html) -Detect people and objects in an image: police review a large photo gallery for a missing person | Object detection and classification | Image | [Object Detection - MXNet](./object-detection.html), [Object Detection - TensorFlow](./object-detection-tensorflow.html) -Tag every pixel of an image individually with a category: self-driving cars prepare to identify objects in their way | Computer vision | Image | [Semantic Segmentation Algorithm](./semantic-segmentation.html) +[Pre-trained models and pre-built solution templates](https://docs.aws.amazon.com/sagemaker/latest/dg/studio-jumpstart.html) | Image Classification Tabular Classification Tabular Regression Text Classification Object Detection Text Embedding Question Answering Sentence Pair Classification Image Embedding Named Entity Recognition Instance Segmentation Text Generation Text Summarization Semantic Segmentation Machine Translation | Here a few examples out of the 15 problem types that can be addressed by the pre-trained models and pre-built solution templates provided by Amazon SageMaker JumpStart: Question answering: chatbot that outputs an answer for a given question. Text analysis: analyze texts from models specific to an industry domain such as finance. | Image, Text, Tabular | Popular models, including Mobilenet, YOLO, Faster R-CNN, BERT, lightGBM, and CatBoostFor a list of pre-trained models available, see [JumpStart Models](https://docs.aws.amazon.com/sagemaker/latest/dg/studio-jumpstart.html#jumpstart-models). For a list of pre-built solution templates available, see [JumpStart Solutions](https://docs.aws.amazon.com/sagemaker/latest/dg/studio-jumpstart.html#jumpstart-solutions). +Supervised learning | Binary/multi-class classification | Predict if an item belongs to a category: an email spam filter | Tabular | [AutoGluon-Tabular](./autogluon-tabular.html), [CatBoost](./catboost.html), [Factorization Machines Algorithm](./fact-machines.html), [K-Nearest Neighbors (k-NN) Algorithm](./k-nearest-neighbors.html), [LightGBM](./lightgbm.html), [Linear Learner Algorithm](./linear-learner.html), [TabTransformer](./tabtransformer.html), [XGBoost algorithm with Amazon SageMaker AI](./xgboost.html) +Regression | Predict a numeric/continuous value: estimate the value of a house | Tabular | [AutoGluon-Tabular](./autogluon-tabular.html), [CatBoost](./catboost.html), [Factorization Machines Algorithm](./fact-machines.html), [K-Nearest Neighbors (k-NN) Algorithm](./k-nearest-neighbors.html), [LightGBM](./lightgbm.html), [Linear Learner Algorithm](./linear-learner.html), [TabTransformer](./tabtransformer.html), [XGBoost algorithm with Amazon SageMaker AI](./xgboost.html) +Time-series forecasting | Based on historical data for a behavior, predict future behavior: predict sales on a new product based on previous sales data. | Tabular | [Use the SageMaker AI DeepAR forecasting algorithm](./deepar.html) +Embeddings: convert high-dimensional objects into low-dimensional space. | Improve the data embeddings of the high-dimensional objects: identify duplicate support tickets or find the correct routing based on similarity of text in the tickets | Tabular | [Object2Vec Algorithm](./object2vec.html) +Unsupervised learning | Feature engineering: dimensionality reduction | Drop those columns from a dataset that have a weak relation with the label/target variable: the color of a car when predicting its mileage. | Tabular | [Principal Component Analysis (PCA) Algorithm](./pca.html) +Anomaly detection | Detect abnormal behavior in application: spot when an IoT sensor is sending abnormal readings | Tabular | [Random Cut Forest (RCF) Algorithm](./randomcutforest.html) +IP anomaly detection | Protect your application from suspicious users: detect if an IP address accessing a service might be from a bad actor | Tabular | [IP Insights](./ip-insights.html) +Clustering or grouping | Group similar objects/data together: find high-, medium-, and low-spending customers from their transaction histories | Tabular | [K-Means Algorithm](./k-means.html) +Topic modeling | Organize a set of documents into topics (not known in advance): tag a document as belonging to a medical category based on the terms used in the document. | Text | [Latent Dirichlet Allocation (LDA) Algorithm](./lda.html), [Neural Topic Model (NTM) Algorithm](./ntm.html) +Textual analysis | Text classification | Assign pre-defined categories to documents in a corpus: categorize books in a library into academic disciplines | Text | [BlazingText algorithm](./blazingtext.html), [Text Classification - TensorFlow](./text-classification-tensorflow.html) +Machine translationalgorithm | Convert text from one language to other: Spanish to English | Text | [Sequence-to-Sequence Algorithm](./seq-2-seq.html) +Text summarization | Summarize a long text corpus: an abstract for a research paper | Text | [Sequence-to-Sequence Algorithm](./seq-2-seq.html) +Speech-to-text | Convert audio files to text: transcribe call center conversations for further analysis | Text | [Sequence-to-Sequence Algorithm](./seq-2-seq.html) +Image processing | Image and multi-label classification | Label/tag an image based on the content of the image: alerts about adult content in an image | Image | [Image Classification - MXNet](./image-classification.html) +Image classification | Classify something in an image using transfer learning. | Image | [Image Classification - TensorFlow](./image-classification-tensorflow.html) +Object detection and classification | Detect people and objects in an image: police review a large photo gallery for a missing person | Image | [Object Detection - MXNet](./object-detection.html), [Object Detection - TensorFlow](./object-detection-tensorflow.html) +Computer vision | Tag every pixel of an image individually with a category: self-driving cars prepare to identify objects in their way | Image | [Semantic Segmentation Algorithm](./semantic-segmentation.html) @@ -64 +64 @@ The following sections provide additional guidance for the Amazon SageMaker AI b -SageMaker JumpStart provides a wide range of pre-trained models, pre-built solution templates, and examples for popular problem types. These use the SageMaker SDK as well as Studio Classic. For more information about these models, solutions, and the example notebooks provided by SageMaker JumpStart, see [SageMaker JumpStart pretrained models](./studio-jumpstart.html). +Amazon SageMaker JumpStart provides a wide range of pre-trained models, pre-built solution templates, and examples for popular problem types. These use the SageMaker SDK as well as Studio Classic. For more information about these models, solutions, and the example notebooks provided by Amazon SageMaker JumpStart, see [SageMaker JumpStart pretrained models](./studio-jumpstart.html).