Processing 100+ million images per month with deep learning
summary
Describes using deep learning to process over 100 million images per month at Linkfluence, focusing on a near real-time deployment architecture. Covers specific system design choices, common pitfalls, and practical tips and tricks for scaling image-processing models.
AI-written summary · about
chapters (6)
- 0:00Start
- 0:44Context: constraints
- 1:08Architecture : kafka based
- 2:20Architecture : monitoring
- 2:52Monitoring models
- 3:39Architecture : models
related talks
20mTraining a foundation image model and deploying it to millions of users
34mDeep Learning at Scale
18mDeep learning for Space imagery applications
45mIntroduction To Deep Learning
40mHow to Implement a Production Ready Deep Learning Model in 12 Weeks
37mAn Initial Look at Deep Learning IO Performance
24mMachine Intelligence gets real
58mDeep Learning in Real Life