sagemaker-python-sdk  by aws

Python SDK for SageMaker model training and deployment

Created 7 years ago
2,190 stars

Top 20.7% on SourcePulse

GitHubView on GitHub
Project Summary

This library provides a Python interface for training and deploying machine learning models on Amazon SageMaker. It targets data scientists and ML engineers who want to leverage AWS's managed infrastructure for their ML workflows, offering streamlined integration with SageMaker's capabilities.

How It Works

The SDK abstracts the complexities of SageMaker's underlying infrastructure, allowing users to define training jobs, deploy models to endpoints, and manage ML resources using familiar Python objects. It supports popular frameworks like TensorFlow, MXNet, PyTorch, and scikit-learn, as well as Amazon's built-in algorithms and custom Docker containers. This approach simplifies the ML lifecycle on AWS by providing a high-level API that maps directly to SageMaker services.

Quick Start & Requirements

  • Install via pip: pip install sagemaker
  • Supported Python versions: 3.9, 3.10, 3.11, 3.12.
  • Requires AWS account credentials and appropriate IAM permissions.
  • Documentation: https://sagemaker.readthedocs.io/en/stable/

Highlighted Details

  • Supports training and deployment for MXNet, TensorFlow, PyTorch, scikit-learn, XGBoost, and Amazon's built-in algorithms.
  • Enables BYO Docker containers for custom training and inference logic.
  • Includes features for automatic model tuning, batch transform, model monitoring, and debugging.
  • Offers SageMaker SparkML Serving for deploying models serialized with MLeap.

Maintenance & Community

  • Developed and maintained by AWS.
  • CI health checks are performed via GitHub Actions.
  • Community support channels are not explicitly mentioned in the README.

Licensing & Compatibility

  • Licensed under the Apache 2.0 License.
  • Compatible with commercial use and closed-source applications.

Limitations & Caveats

  • Integration tests require specific AWS IAM roles and potentially ECR repository setup.
  • Telemetry is enabled by default but can be opted out.
Health Check
Last Commit

22 hours ago

Responsiveness

1 week

Pull Requests (30d)
16
Issues (30d)
4
Star History
9 stars in the last 30 days

Explore Similar Projects

Starred by Amanpreet Singh Amanpreet Singh(Cofounder of Contextual AI), Yineng Zhang Yineng Zhang(Inference Lead at SGLang; Research Scientist at Together AI), and
7 more.

truss by basetenlabs

0.2%
1k
Model deployment tool for productionizing AI/ML models
Created 3 years ago
Updated 1 day ago
Starred by Jiayi Pan Jiayi Pan(Author of SWE-Gym; MTS at xAI), Christian Laforte Christian Laforte(Distinguished Engineer at NVIDIA; Former CTO at Stability AI), and
3 more.

lightning-hydra-template by ashleve

0.1%
5k
ML experimentation template using PyTorch Lightning + Hydra
Created 4 years ago
Updated 1 year ago
Starred by Chip Huyen Chip Huyen(Author of "AI Engineering", "Designing Machine Learning Systems"), Travis Fischer Travis Fischer(Founder of Agentic), and
2 more.

modelscope by modelscope

0.2%
8k
Model-as-a-Service library for model inference, training, and evaluation
Created 3 years ago
Updated 1 day ago
Starred by Peter Norvig Peter Norvig(Author of "Artificial Intelligence: A Modern Approach"; Research Director at Google), Aravind Srinivas Aravind Srinivas(Cofounder of Perplexity), and
98 more.

tensorflow by tensorflow

0.1%
192k
Open-source ML framework
Created 10 years ago
Updated 12 hours ago
Feedback? Help us improve.