Amazon SageMaker Built-in algorithms and Spot checkpointing

0

Which Amazon SageMaker built-in algorithms support checkpointing? In the documentation it says that:

SageMaker built-in algorithms and marketplace algorithms that do not checkpoint are currently limited to a MaxWaitTimeInSeconds of 3600 seconds (60 minutes).

However, in the algorithms I don't find any pointer to "checkpoint" or "spot". Can you help me out?

AWS
demandé il y a 4 ans351 vues
1 réponse
0
Réponse acceptée

This is the best resource that I've found to clarify this:

https://aws.amazon.com/blogs/aws/managed-spot-training-save-up-to-90-on-your-amazon-sagemaker-training-jobs/

Built-in algorithms: computer vision algorithms support checkpointing (Object Detection, Semantic Segmentation, and very soon Image Classification). As they tend to train on large data sets and run for longer than other algorithms, they have a higher likelihood of being interrupted. Other built-in algorithms do not support checkpointing for now.

Also:

Please note that TensorFlow uses checkpoints by default. For other frameworks, you’ll find examples in our sample notebooks and in the documentation.

répondu il y a 4 ans

Vous n'êtes pas connecté. Se connecter pour publier une réponse.

Une bonne réponse répond clairement à la question, contient des commentaires constructifs et encourage le développement professionnel de la personne qui pose la question.

Instructions pour répondre aux questions