Checkpoints in Amazon SageMaker
Use checkpoints in Amazon SageMaker to save the state of machine learning (ML) models during training. Checkpoints are snapshots of the model and can be configured by the callback functions of ML frameworks. You can use the saved checkpoints to restart a training job from the last saved checkpoint.
Using checkpoints, you can do the following:
-
Save your model snapshots under training due to an unexpected interruption to the training job or instance.
-
Resume training the model in the future from a checkpoint.
-
Analyze the model at intermediate stages of training.
-
Use checkpoints with S3 Express One Zone for increased access speeds.
-
Use checkpoints with SageMaker managed spot training to save on training costs.
The SageMaker training mechanism uses training containers on Amazon EC2 instances, and the
checkpoint files are saved under a local directory of the containers (the default is
/opt/ml/checkpoints). SageMaker provides the functionality to copy the
checkpoints from the local path to Amazon S3 and automatically syncs the checkpoints in that
directory with S3. Existing checkpoints in S3 are written to the SageMaker
container at the start of the job, enabling jobs to resume from a checkpoint. Checkpoints
added to the S3 folder after the job has started are not copied to the training
container. SageMaker also writes new checkpoints from the container to S3 during training.
If a checkpoint is deleted in the SageMaker container, it will also be deleted in the S3
folder.
You can use checkpoints in Amazon SageMaker with the Amazon S3 Express One Zone storage class (S3 Express One Zone) for faster access to checkpoints. When you enable checkpointing and specify the S3 URI for your checkpoint storage destination, you can provide an S3 URI for a folder in either an S3 general purpose bucket or an S3 directory bucket. S3 directory buckets that are integrated with SageMaker can only be encrypted with server-side encryption with Amazon S3 managed keys (SSE-S3). Server-side encryption with AWS KMS keys (SSE-KMS) is not currently supported. For more information on S3 Express One Zone and S3 directory buckets, see What is S3 Express One Zone.
If you are using checkpoints with SageMaker managed spot training, SageMaker manages checkpointing your model training on a spot instance and resuming the training job on the next spot instance. With SageMaker managed spot training, you can significantly reduce the billable time for training ML models. For more information, see Managed Spot Training in Amazon SageMaker.
Topics
Checkpoints for frameworks and algorithms in SageMaker
Use checkpoints to save snapshots of ML models built on your preferred frameworks within SageMaker.
SageMaker frameworks and algorithms that support checkpointing
SageMaker supports checkpointing for AWS Deep Learning Containers and a subset of
built-in algorithms without requiring training script changes. SageMaker saves the
checkpoints to the default local path '/opt/ml/checkpoints' and copies them
to Amazon S3.
-
Deep Learning Containers: TensorFlow
, PyTorch , MXNet , and HuggingFace Note
If you are using the HuggingFace framework estimator, you need to specify a checkpoint output path through hyperparameters. For more information, see Run training on Amazon SageMaker
in the HuggingFace documentation. -
Built-in algorithms: Image Classification, Object Detection, Semantic Segmentation, and XGBoost (0.90-1 or later)
Note
If you are using the XGBoost algorithm in framework mode (script mode), you need to bring an XGBoost training script with checkpointing that's manually configured. For more information about the XGBoost training methods to save model snapshots, see Training XGBoost
in the XGBoost Python SDK documentation.
If a pre-built algorithm that does not support checkpointing is used in a managed spot training job, SageMaker does not allow a maximum wait time greater than an hour for the job in order to limit wasted training time from interrupts.
For custom training containers and other frameworks
If you are using your own training containers, training scripts, or other frameworks
not listed in the previous section, you must properly set up your training script using
callbacks or training APIs to save checkpoints to the local path
('/opt/ml/checkpoints') and load from the local path in your training
script. SageMaker estimators can sync up with the local path and save the checkpoints to
Amazon S3.
Considerations for checkpointing
Consider the following when using checkpoints in SageMaker.
-
To avoid overwrites in distributed training with multiple instances, you must manually configure the checkpoint file names and paths in your training script. The high-level SageMaker checkpoint configuration specifies a single Amazon S3 location without additional suffixes or prefixes to tag checkpoints from multiple instances.
-
The SageMaker Python SDK does not support high-level configuration for checkpointing frequency. To control the checkpointing frequency, modify your training script using the framework's model save functions or checkpoint callbacks.
-
If you use SageMaker checkpoints with SageMaker Debugger and SageMaker distributed and are facing issues, see the following pages for troubleshooting and considerations.
