How We Built a Stable Regression Environment for Over 1,000 Services

This article details the challenges, planning, and step‑by‑step execution of creating an isolated regression environment for a large micro‑service platform, covering quantifiable goals, solution evaluation, infrastructure refactoring, data migration, deployment results, and future improvements.

Infrastructure Automationcloud migrationenvironment governance

0 likes · 14 min read

How We Built a Stable Regression Environment for Over 1,000 Services

360 Quality & Efficiency

Jan 17, 2020 · Artificial Intelligence

File Release Application Prediction Model Using GBDT

This article describes how a GBDT‑based prediction model was built to forecast file release application parameters such as volume ratio, target audience, and gray level, covering data collection, feature engineering, model training, service deployment, and practical considerations for handling bad cases.

GBDTdata preprocessingfile release

0 likes · 8 min read

File Release Application Prediction Model Using GBDT