Deep Learning with TensorFlow, PyTorch and MXNet on AWS
summary
How to run large-scale deep learning on AWS using TensorFlow, PyTorch and Apache MXNet: scaling distributed training across many GPUs, reading massive training datasets, and choosing the right EC2 instance types (GPU vs CPU). Also covers deploying models for low-latency, secure, production inference and cost-optimization techniques for distributed training and inference.
AI-written summary · about
chapters (11)
- 0:00Welcome
- 1:27ML vs DL
- 3:57DL project challenges
- 7:13AWS DL infrastructure
- 10:07Benchmarks and optimizations
- 15:04SageMaker training
- 18:20Large dataset handling
- 20:49Instance selection
- 22:44Inference and microservices
- 24:53Framework optimizations
- 29:41Summary and resources
related talks
01h 16mDeep Dive into Deep Learning with MXNet
54mDeep learning on AWS with NVIDIA: From training to deployment
43mDeep learning beyond the learning
19mDeveloping large-scale machine learning algorithms on Amazon SageMaker
23mBest Practices for Building Production-Grade Deep Learning Systems
42mAlex Smola, Professor in the Machine Learning Department, Carnegie Mellon University
35mLarge Scale Distributed Deep Learning on Kubernetes Clusters
31mBest Practices for Producing High-Quality Deep Learning Systems