aws / aws/sagemaker-python-sdk

How to change default values in constraints generated by baselines (`sagemaker.model_monitor.***.suggest_baseline()`)

Ouverte
#2,628 0 commentaires 1 réaction 1 personne assignée Réclamée par @rsareddy0329 Voir sur GitHub
type: documentation type: question
Langage dominant
Python
Étoiles
2.3k
Forks
1.3k
Merge moyen
1 j 22 h
PR mergées (30 j)
35

Description

**What did you find confusing? Please describe.**
I have failed to find any documentation that let me change the default values in constraints.json that
[`sagemaker.model_monitor.DefaultModelMonitor.suggest_baseline()`](https://sagemaker.readthedocs.io/en/stable/api/inference/model_monitor.html?highlight=DefaultModelMonitor#sagemaker.model_monitor.model_monitoring.DefaultModelMonitor.suggest_baseline) saves to S3 after a processing job. one of the constraints that I would like to change are `domain_content_threshold` from `0.1` to `0.25`.

**Describe how documentation can be improved**
Provide documentation to change the default values of multiple constraints from [constraints.json file](https://docs.aws.amazon.com/sagemaker/latest/dg/model-monitor-byoc-constraints.html) in documentation for each of the `suggest_baseline` methods (I meant methods from `sagemaker.model_monitor.***`). And also let users use custom code(or a python function) for `comparison_method` in `distribution_constraints`. Having to bring custom container for one modification from sagemaker defaults is not optimal and time consuming for users to build. Moreover, sagemaker's model-analyzer images are not public anyway (or I haven't been able to pull pull them to local docker) for users to able to improve upon or change couple of things to suit their needs.

**Additional context**
Add any other context or screenshots about the documentation request here.

Guide de contribution

Ouvrir le guide de contribution

Évaluation

Cette issue n'a pas encore été évaluée.

Recevez les nouvelles issues par e-mail

Un résumé court des issues GitHub adaptées aux débutants.