简体   繁体   English

我正在 sagemaker 中进行 model 偏差监控,是否可以手动运行监控作业?

[英]I'm working on model bias monitoring in sagemaker, Is it possible to run the monitoring job manually?

I've done the baseline creation part where I'm getting analysis_config.json file.我已经完成了获取analysis_config.json文件的基线创建部分。 So next step is doing the monitoring job part, Is it possible to run monitoring job manually for Monitor Bias Drift and the next step Feature Attribution Drift ?所以下一步是做监控工作部分,是否可以为Monitor Bias Drift和下一步Feature Attribution Drift手动运行监控工作?

Yes you can run the model monitoring job manually by configuring the process job with appropriate parameters.是的,您可以通过使用适当的参数配置流程作业来手动运行 model 监控作业。 Please find an example in the reference here请在此处的参考中找到一个示例

暂无
暂无

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

相关问题 SageMaker:如何手动运行 Model 质量监控作业? - SageMaker: How to manually run a Model Quality Monitor Job? Sagemaker:如何调试Model监控(数据质量和model质量)? - Sagemaker: How to debug Model monitoring(data quality and model quality)? Sagemaker Model 执行 latest_execution.describe() 时出现监控错误 - Sagemaker Model Monitoring error while performing latest_execution.describe() 是否有任何可能的方法可以在没有端点的情况下在 aws 中进行 model 监控 - Is there any possible ways to do model monitoring in aws without an endpoint Celery 监控不适用于数据库代理 - Celery monitoring not working with database broker 使用pytest,单元测试是否可以知道正在运行的代码覆盖率监视处于运行状态? - Using pytest, is it possible for a unit test to know that it is being run with code coverage monitoring on? 我在我的python代码上显示输出以进行网络监控时遇到问题。 不显示结果但没有错误 - I'm having trouble displaying output on my python code for network monitoring. Not displaying result but no error 是否可以在不使用 SageMaker SDK 的情况下为我在 AWS SageMaker 中创建的模型设置终端节点 - Is it possible set up an endpoint for a model I created in AWS SageMaker without using the SageMaker SDK Flask 监控仪表板不适用于 after_request - Flask Monitoring Dashboard not working with after_request 雷击监控 - Thunderpush monitoring
 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM