Deep Learning with the GPUs in Production
summary
Covers using GPUs for deep learning in production, focusing on practical tools and libraries (such as those from Skymind) to deploy neural network models at scale. Explains challenges around GPU-based training and inference in production environments and approaches to integrate GPU workflows into developer pipelines.
AI-written summary · about
chapters (14)
- 0:00Start
- 1:05skymind | WHO WE ARE
- 2:45skymind | DEFINING PRODUCTION
- 6:23skymind VARIOUS KINDS OF GPU CLUSTERS
- 9:14skymind | BIG DATA HYBRID CLUSTER
- 9:28skymind | DEEPLEARNING4J (DL4J)
- 16:21skymind TRAINING UI
- 17:04skymind | TRAINING UI
- 19:10skymind | INFERENCE
- 23:32skymind PROBLEM TO THINK ABOUT JOBS ON GPU CLUSTER
- 30:05skymind | KEY FOR BIG DATA CLUSTERS
- 31:34skymind | WORKFLOW FOR DATA SCIENTISTS
- 34:37skymind | WORKFLOW FOR DATA ENGINEERS AND MICROSERVICES DEVELOPERS
- 37:38skymind RECENT CASE STUDY: NASA JPL
related talks
16mDeep Learning in Production Seen by a Data Engineer
44mDeep Learning in Production
36mProduction GPU Cluster with K8s for AI and DL Workloads
51mDeploying Deep Learning models on Kubernetes clusters with GPU Enabled
23mBest Practices for Building Production-Grade Deep Learning Systems
33mOvercoming the Barriers to Deep Learning in Production
31mBest Practices for Producing High-Quality Deep Learning Systems
26mDeep learning in Go - Present and future