aws/sagemaker-training-toolkit

View on GitHub

Train machine learning models within a 🐳 Docker container using 🧠 Amazon SageMaker.

Stars: 530
Forks: 140
Open beginner issues: 0
Indexed issues: 51
Avg merge: 51m
Merged PRs (30d): 4

Dominant language: Python
License: Apache-2.0
Last GitHub push: Sep 16, 2026
Latest indexed: Sep 14, 2026
Contributing guide: Contributing guide
Code of conduct: Code of conduct
Beginner labels: No beginner labels indexed

Beginner friendly Everything

Open All

51 open issues indexed

Inquiry about Python 3.11, 3.12, and 3.13 support and testing Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#236 · 2 comments · 6 reactions · Opened Jul 16, 2025
Loosen or update protobuf pinned version Open

Difficulty 2/5 1-3 hours Newbie friendliness 45/100

aws/sagemaker-training-toolkit#234 · Opened May 19, 2025
setuptools 78.0.1 incompatibility Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#233 · Opened Mar 25, 2025
Silent Failure if custom image puts something into /opt/ml/code Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#222 · 1 comment · 1 reaction · Opened Sep 5, 2024
SageMaker training toolkit reorders hyperparameters Open

Difficulty 2/5 1-3 hours Newbie friendliness 45/100

aws/sagemaker-training-toolkit#221 · Opened Sep 5, 2024
Add 'ml.p5.48xlarge' as a supported instance for SM_EFA_NCCL_INSTANCES. Open

Difficulty 2/5 1-3 hours Newbie friendliness 48/100

aws/sagemaker-training-toolkit#219 · 1 comment · Opened Sep 4, 2024
Extend documentation regarding distributed training for own Docker containers. Open

Difficulty 2/5 1-3 hours Newbie friendliness 38/100

aws/sagemaker-training-toolkit#218 · Opened Aug 26, 2024
Get region with ENV var Open

Difficulty 2/5 1-3 hours Newbie friendliness 35/100

aws/sagemaker-training-toolkit#207 · Opened Feb 8, 2024
Invalid dash-separated options for description-file Open

Difficulty 1/5 Under an hour Newbie friendliness 35/100

aws/sagemaker-training-toolkit#206 · 1 reaction · Opened Dec 20, 2023
Training Job "Successful" despite failing due to 100% disk usage Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#204 · Opened Nov 8, 2023
ModuleNotFoundError: Sagemaker only copies entry_point file to /opt/ml/code/ instead of the holy-cloned source code Open

Difficulty 3/5 1-2 days Newbie friendliness 45/100

aws/sagemaker-training-toolkit#200 · Opened Oct 15, 2023
Issue when training in local mode with huggingface training container Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#193 · 1 comment · Opened Sep 11, 2023
P5 instance support Open

Difficulty 2/5 1-3 hours Newbie friendliness 35/100

aws/sagemaker-training-toolkit#191 · Opened Aug 31, 2023
Adding sys.path to PYTHONPATH breaks virtual environments Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#186 · Opened Jun 29, 2023
Deepspeed Launcher Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

aws/sagemaker-training-toolkit#184 · 1 reaction · Opened May 26, 2023
Publish wheels to PyPI Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#174 · 7 reactions · Opened Feb 15, 2023
Passing SIGTERM to entrypoint to be able to handle SPOT failures gracefully in user-code Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#173 · Opened Jan 31, 2023
Remove magic strings for attributes like instance type Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

aws/sagemaker-training-toolkit#164 · 1 reaction · Opened Dec 5, 2022
Python 3.6 unsupported [bug/question] Open

Difficulty 3/5 1-2 days Newbie friendliness 45/100

aws/sagemaker-training-toolkit#161 · 1 comment · Opened Nov 15, 2022
Mpi mode sets all nodes to the same SM_CURRENT_HOST Open

Difficulty 2/5 1-3 hours Newbie friendliness 45/100

aws/sagemaker-training-toolkit#158 · Opened Oct 31, 2022
No support for Python 3.10 Open

Difficulty 2/5 1-3 hours Newbie friendliness 48/100

aws/sagemaker-training-toolkit#129 · 9 comments · 2 reactions · Opened May 31, 2022
Hyperparameters not shell escaped Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#128 · Opened May 23, 2022
Pass SIGTERM to training script to stop training Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#125 · 2 reactions · Opened May 19, 2022
Custom_Overrides Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

aws/sagemaker-training-toolkit#112 · Opened Dec 10, 2021
Unable to install sagemaker-training on Windows Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#110 · 4 comments · 7 reactions · Opened Oct 21, 2021
How does sagemaker-training-toolkit complement sagemaker-python-sdk? Open

Difficulty 4/5 3-5 days Newbie friendliness 38/100

aws/sagemaker-training-toolkit#105 · 2 reactions · Opened Apr 22, 2021
Don't set `sagemaker_s3_output` via hyperparameter Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

aws/sagemaker-training-toolkit#104 · 2 reactions · Opened Apr 21, 2021
Add gcc package requirment Open

Difficulty 3/5 1-2 days Newbie friendliness 38/100

aws/sagemaker-training-toolkit#94 · 11 reactions · Opened Jan 20, 2021
how to pass predictor script for running batch transformers. Open

Difficulty 3/5 1-2 days Newbie friendliness 25/100

aws/sagemaker-training-toolkit#93 · Opened Jan 14, 2021
SageMaker Endpoint stuck at “Creating” Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

aws/sagemaker-training-toolkit#92 · Opened Jan 12, 2021
Sagemaker Fails to download code from S3 Open

type: question

Difficulty 2/5 1-3 hours Newbie friendliness 45/100

aws/sagemaker-training-toolkit#85 · Opened Nov 26, 2020
Which aws service is the most suitable/used to launch a scheduled training job? Open

type: question

Difficulty 3/5 1-2 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#84 · Opened Nov 24, 2020
Enable functional test for mpi Open

type: maintenance

Difficulty 3/5 1-2 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#83 · Opened Nov 1, 2020
Model and output files do not get saved to S3 when training own model Open

type: question

Difficulty 4/5 3-5 days Newbie friendliness 32/100

aws/sagemaker-training-toolkit#82 · 5 comments · 1 reaction · Opened Oct 6, 2020
How to save non-model artifacts from a container (output_data_dir) Open

type: question

Difficulty 3/5 1-2 days Newbie friendliness 43/100

aws/sagemaker-training-toolkit#80 · 2 comments · 3 reactions · Opened Sep 29, 2020
Enhance UX for training Open

type: enhancement type: UX

Difficulty 5/5 Over a week Newbie friendliness 20/100

aws/sagemaker-training-toolkit#77 · 2 reactions · Opened Aug 6, 2020
Unable to run a Tensorflow Estimator. Open

type: question

Difficulty 4/5 3-5 days Newbie friendliness 25/100

aws/sagemaker-training-toolkit#72 · 3 comments · Opened Jul 21, 2020
Custom MPI options doesn't override the flags Open

type: bug

Difficulty 2/5 1-3 hours Newbie friendliness 45/100

aws/sagemaker-training-toolkit#70 · 3 comments · 1 reaction · Opened Jul 7, 2020
Pass args to training script entrypoint for MPI-based Distributed training Open

type: enhancement

Difficulty 4/5 3-5 days Newbie friendliness 30/100

aws/sagemaker-training-toolkit#69 · 3 comments · Opened Jul 1, 2020
Failed to parse string hyperparameter Open

type: question

Difficulty 3/5 1-2 days Newbie friendliness 25/100

aws/sagemaker-training-toolkit#66 · 6 comments · Opened Jun 26, 2020
Bash Command ENTRYPOINT Expects `train` argument Open

type: question

Difficulty 4/5 3-5 days Newbie friendliness 30/100

aws/sagemaker-training-toolkit#65 · 8 comments · 1 reaction · Opened Jun 23, 2020
Entry point package doesn't seem to work with nested directories Open

type: bug

aws/sagemaker-training-toolkit#54 · 18 comments · 1 reaction · 1 assignee · Opened Feb 5, 2020
sagemaker container doesn't support running container image with more privilege Open

type: enhancement

Difficulty 5/5 Over a week Newbie friendliness 25/100

aws/sagemaker-training-toolkit#21 · 1 comment · Opened Jan 6, 2020
Pytorch Sagemaker Container STDERR output Open

priority: high type: bug

Difficulty 3/5 1-2 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#37 · 5 comments · Opened Oct 22, 2019
improve the process of reading data in distributed training Open

priority: medium type: enhancement

Difficulty 5/5 Over a week Newbie friendliness 25/100

aws/sagemaker-training-toolkit#17 · Opened Sep 24, 2019
consider using Horovod for distributed training Open

priority: medium related: Horovod type: enhancement

Difficulty 5/5 Over a week Newbie friendliness 25/100

aws/sagemaker-training-toolkit#16 · Opened Sep 24, 2019
improve MPI error output Open

type: UX

Difficulty 4/5 3-5 days Newbie friendliness 25/100

aws/sagemaker-training-toolkit#5 · Opened Sep 24, 2019
add documentation to explain MPI settings Open

priority: low type: documentation

Difficulty 3/5 1-2 days Newbie friendliness 35/100

aws/sagemaker-training-toolkit#15 · Opened Sep 24, 2019
consider a better default Request content type Open

priority: low type: enhancement

Difficulty 4/5 3-5 days Newbie friendliness 25/100

aws/sagemaker-training-toolkit#14 · Opened Sep 24, 2019
add error handling for importing framework module Open

type: UX

Difficulty 2/5 1-3 hours Newbie friendliness 52/100

aws/sagemaker-training-toolkit#4 · Opened Sep 24, 2019

aws/sagemaker-training-toolkit

Get new issues in your inbox