Watch a 17-minute conference talk by Google Principal Engineer Cormac Brick exploring the landscape of machine learning software ecosystems and their future trajectory. Gain insights into prominent frameworks like JAX, PyTorch, TensorFlow, and Keras, while discovering the role of ML compilers such as OpenXLA and open source models including Gemma and Llama. Explore performance considerations across CPU, GPU, and TPU architectures, along with edge computing and on-device deployment challenges. Understand the critical hardware requirements for achieving breakthrough results, including system resilience and mathematical format considerations that drive the next generation of ML software development.
Machine Learning Software Ecosystems: Frameworks, Compilers, and Models - Part 1
Open Compute Project via YouTube
Overview
Syllabus
ML Software Ecosystems 1
Taught by
Open Compute Project
Related Courses
-
Production Machine Learning Systems
1.0 -
Deep Learning
-
Fine-tuning Vision Language Models with PaLiGemma-3B - A Practical Guide
-
Accelerate PyTorch Workloads with PyTorch/XLA
-
Developer Transition: Machine Learning to Quantum Machine Learning
-
How to Pick a GPU and Inference Engine for Large Language Models
