Meet the Expert Dean Wampler on Scaling MLAI Applications with Ray

MP4 | Video: h264, 1920x1080 | Audio: AAC, 44.1 KHz, 2 Ch
Genre: eLearning | Language: English | Duration: 57m | Size: 1.76 GB
Modern ML and AI applications require a lot of compute power, which usually means distribution over a cluster of machines, as well as management of distributed state, such as the model parameters being trained. Ray, a high-performance distributed execution framework developed by UC Berkeley's RISELab, is targeted at large-scale machine learning and reinforcement learning applications. Ray's features make it suitable for any Python-based application that needs cluster-wide scalability.