---
id: sagemaker-train
version: "1.19.0"
license: unclear
license_treatment: unclear
maintenance: active
---

# sagemaker-train — Open source library for training and deploying models on Amazon SageMaker.

License: unclear · Maintenance: active · Downloads: 1.3M/mo

## What it is and what it does

sagemaker-train is the AWS SageMaker Python SDK's training module, a library for programmatically training and deploying machine learning models on Amazon's managed SageMaker service. It abstracts the complexity of distributed training, GPU provisioning, and model hosting by providing a Python interface to SageMaker's capabilities. The package supports training with popular frameworks like Apache MXNet and TensorFlow, Amazon's optimized built-in algorithms, and custom algorithms packaged in SageMaker-compatible Docker containers.

The library is actively maintained, tested on Python 3.10–3.12, and depends on a stack of 12 runtime packages including sagemaker-core, mlflow, pydantic, and paramiko. It includes telemetry by default (opt-out available via configuration) and requires AWS credentials and appropriate IAM permissions to function. The package is in Alpha status and is designed for developers building ML workflows on AWS infrastructure.

Use it for:

- Train deep learning models using TensorFlow or Apache MXNet on SageMaker's managed infrastructure without managing compute directly.
- Deploy trained models to SageMaker endpoints for real-time inference or batch predictions.
- Use Amazon's pre-built, optimized algorithms for common ML tasks like classification or regression.
- Integrate custom training code packaged in Docker containers with SageMaker's distributed training and monitoring.
- Track and manage ML experiments using MLflow integration for reproducibility and model comparison.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Trains and deploys machine learning models on Amazon SageMaker using popular frameworks like Apache MXNet, TensorFlow, and Amazon's built-in algorithms, or custom Docker containers.

Yes, if you are building ML workflows on AWS and want a managed training and deployment platform. The package is actively maintained, has low install friction, and integrates well with SageMaker's ecosystem. However, verify the license terms (Apache 2.0 is mentioned in the description but not confirmed in metadata) and ensure your AWS account has the necessary IAM permissions before relying on it. Not suitable if you need to train models outside AWS or prefer a cloud-agnostic ML framework.

## Install

pip install sagemaker-train
uv add sagemaker-train
poetry add sagemaker-train

## Installing sagemaker-train

Before you install: Low install friction with a pure-Python wheel. Active maintenance with a recent release (3 days old) and steady repository activity. Requires Python 3.10 or later and depends on 12 runtime packages including sagemaker-core, mlflow, and pydantic.

License in practice: License treatment is unclear—the description mentions Apache 2.0 licensing, but the fact sheet contains no SPDX identifier or confirmed license metadata. Verify the actual license terms before relying on this package in a commercial or restricted-license context.

Quickstart:

pip install sagemaker-train

from sagemaker_train import ...
# See https://sagemaker.readthedocs.io for detailed usage examples

Requires Python 3.10, 3.11, or 3.12. AWS credentials and appropriate IAM permissions are needed to interact with SageMaker services.

Verify before relying:

- Whether the Apache 2.0 license mentioned in the description is the actual applied license given the unclear license_treatment in metadata.
- Specific API surface and which frameworks (MXNet, TensorFlow, PyTorch, HuggingFace) are fully supported in this 1.19.0 release.
- Whether telemetry can be fully disabled and what data is collected before opt-out configuration is applied.

## Package facts

- License: not declared (unclear)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 1.3M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

sagemaker model training, aws ml deployment sdk, tensorflow training on sagemaker, amazon sagemaker python, distributed ml training aws, sagemaker model hosting, mxnet sagemaker training, aws-sagemaker, ml-training, distributed-computing

[View on SkillFed](https://skillfed.io/packages/sagemaker-train) · [View on PyPI](https://pypi.org/project/sagemaker-train/)