NannyML vs STAMM

NannyML monitors deployed machine-learning models and estimates their performance when ground truth is delayed or unavailable. It detects concept drift as well as multivariate and univariate data drift, and connects alerts to changes in model performance. Data-quality checks and ranked drift alerts can help teams investigate features associated with performance issues. Users can relate model performance to monetary or other business outcomes. NannyML Cloud supports webhooks for retraining actions and an SDK for automating monitoring-data ingestion. Cloud is available through Azure and AWS marketplaces as SaaS or a managed application. With the managed-application option, the application and required infrastructure are provisioned within the customer's cloud subscription. For its Azure managed application, NannyML documents encrypted traffic and disks, but says automated backups were still being implemented and advises customers to safeguard model inputs and outputs elsewhere. The open-source library installs with pip or conda. Direct local dataset uploads through the Cloud UI must be smaller than 100 MB. Plans include a free self-managed Open Source option, Starter at $399 a month, and Scale at $999 a month.

STAMM is an open-source MLOps framework for deploying and maintaining machine-learning soft sensors in industrial processes. It monitors live process data for concept drift and changes in operating regimes, and its registry records model versions, configurations, artifacts, validation results, and metadata. A dashboard shows process measurements, soft-sensor outputs, drift signals, and historical context, with support for human-in-the-loop labelling. STAMM serves Python and R soft sensors through REST inference. Its event-driven orchestrator calls the registry over REST and links predictions to the data snapshot that produced them. The drift detector package offers 10 detectors through a common interface and can also be used separately. Data can enter through equipment REST hooks, MQTT applications, or frameworks such as LEAF. InfluxDB is the reference time-series store; a PostgreSQL adapter is described as in progress. The reference Docker Compose installation supports Linux, macOS, or Windows with Docker 24 or later, and 8 GB RAM or more is recommended. Released under Apache License 2.0, STAMM is intended for process modelers, ML engineers, operators, and production managers.