

Episode 549: William Falcon Optimizing Deep Learning Models
11 snips Feb 3, 2023
Chapters
Transcript
Episode notes
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Introduction
00:00 • 2min
The Challenges of Deep Learning
01:37 • 3min
Do You Think You Need Deep Learning?
04:27 • 2min
Optimizing Deep Learning Models
06:45 • 2min
Is Training Versus Inferencing a Thing?
09:15 • 3min
Is There a Transition Point for Distributed Training?
12:06 • 2min
MLOps Tasks - Can You Do Two Iterations a Day?
14:26 • 3min
Is There a Barrier to Entry in the Deep Learning Industry?
17:44 • 3min
How to Train GPUs in the Cloud?
20:56 • 3min
Getting Rid of Boilerplate in PyTorch
23:58 • 4min
Is There Any Pitfall in Deep Learning?
28:15 • 4min
Using Lightning to Train Machine Learning Models
31:55 • 2min
PyTorch - What's the Difference Between a Torch and a Module?
34:12 • 2min
How to Train, Validate, Test and Predict
36:38 • 2min
How to Use Lightning to Improve Your Model Performance
38:53 • 3min
Lightning and Data Versioning
41:52 • 2min
Debugging in Machine Learning?
43:54 • 2min
Debugging Distributed Training Bugs?
45:29 • 2min
Decoupling Data From the Model From the Hardware
47:24 • 2min
Is TensorFlow the Right Approach?
48:55 • 2min
Open-Source and the Needs of Academic Researchers Compared to Commercial Users
51:13 • 2min
Getting Started With PyTorch and Lightning
53:34 • 2min
Lightning - The Last Deep Learning Framework You'll Ever Need?
55:45 • 2min
Developing an Operating System for Artificial Intelligence
57:23 • 2min
The Next Big Thing Is AI, Right?
58:54 • 2min