Aparna DhinakaraninTowards Data ScienceHow to Understand and Use the Jensen-Shannon DivergenceA primer on the math, logic, and pragmatic application of JS Divergence — including how it is best used in drift monitoringMar 2, 20233
Aparna DhinakaraninTowards Data ScienceUnderstanding KL DivergenceA guide to the math, intuition, and practical use of KL divergence — including how it is best used in drift monitoringFeb 2, 20232Feb 2, 20232
Rajesh MoreEnsuring Model Reliability: Building a Consistency Test Pipeline with Kubernetes and Cronjob (GCP)In this post, we will walk you through the design and implementation of a Consistency Test Pipeline. This pipeline helps ensure the…May 27May 27
Aparna DhinakaraninTowards Data ScienceMeasuring Embedding DriftApproaches for measuring embedding/vector drift for unstructured data, including for computer vision and natural language processing modelsDec 8, 20224Dec 8, 20224
Aparna DhinakaraninTowards Data ScienceHow to Understand and Use the Jensen-Shannon DivergenceA primer on the math, logic, and pragmatic application of JS Divergence — including how it is best used in drift monitoringMar 2, 20233
Aparna DhinakaraninTowards Data ScienceUnderstanding KL DivergenceA guide to the math, intuition, and practical use of KL divergence — including how it is best used in drift monitoringFeb 2, 20232
Rajesh MoreEnsuring Model Reliability: Building a Consistency Test Pipeline with Kubernetes and Cronjob (GCP)In this post, we will walk you through the design and implementation of a Consistency Test Pipeline. This pipeline helps ensure the…May 27
Aparna DhinakaraninTowards Data ScienceMeasuring Embedding DriftApproaches for measuring embedding/vector drift for unstructured data, including for computer vision and natural language processing modelsDec 8, 20224
Aparna DhinakaraninTowards Data ScienceDemystifying NDCGHow to best use this important metric for monitoring ranking modelsJan 25, 20235
StanEvaluate Your RAG System Using NDCG“If we cannot measure it, we cannot improve it” - Peter DruckerMay 12
Ansh TanwarWhy you need to babysit ML models after deployment? [part-2]🤖nannyML for post-deployment model monitoringMay 122