Properties (64)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:Cloud_Computing_Service
|
| gptkbp:allows |
Custom algorithm training
|
| gptkbp:enables |
Distributed training
|
| gptkbp:features |
Hyperparameter_optimization
|
| https://www.w3.org/2000/01/rdf-schema#label |
Amazon SageMaker Training
|
| gptkbp:integration |
gptkb:Amazon_S3
gptkb:AWS_Lambda |
| gptkbp:is_accessible_by |
gptkb:AWS_Management_Console
gptkb:AWS_CLI SDKs |
| gptkbp:is_designed_to |
Developers
Data scientists Machine learning engineers |
| gptkbp:is_part_of |
gptkb:Amazon_Web_Services
|
| gptkbp:is_used_in |
Jupyter notebooks
|
| gptkbp:offers |
Automatic model tuning
Model training at scale Model monitoring Training job notifications Model explainability tools Batch transform capabilities Cost-effective training options Model retraining capabilities Preprocessing scripts Training job retries Training job tagging Pre-built_Docker_containers |
| gptkbp:provides |
Security features
Integration with CI/CD tools Built-in data labeling Data preprocessing capabilities Experiment management Managed training environment Training data management Training metrics Training job monitoring Training job scheduling Integration_with_AWS_Glue Integration_with_Amazon_CloudWatch Integration_with_Amazon_ECR Integration_with_SageMaker_Studio |
| gptkbp:supports |
Resource management
Data versioning Custom metrics Data augmentation Feature engineering Model evaluation Frameworks like TensorFlow Data parallelism Real-time inference Model serving Model deployment Built-in algorithms Model versioning Data drift detection Model parallelism Multi-instance training Asynchronous inference Frameworks like MXNet Spot instances for training Custom_Docker_containers Frameworks_like_PyTorch Hyperparameter_tuning_jobs VPC_integration |