ClearML for training neural networks: why it and how we became friends with it
summary
ClearML is presented as a lightweight MLOps solution for small teams that need to train neural networks without heavy maintenance overhead. The talk explains how to use ClearML to organize training environments, build helpful wrappers for easier use, and set up logging and resource management.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 2:28What is ClearML
- 7:12ClearML features
- 9:39Problems before ClearML
- 11:28Solving the problems
- 13:32Automating training with ClearML
- 15:39Resource orchestration
- 17:45Project organization
- 18:50Dataset caching
- 20:37Dataset storage and handling
- 21:40Dataset versioning problems
- 22:43Logging and monitoring
- 24:31Implementation results
- 27:01Process standardization
- 28:25Q&A
related talks
53mClearML: Your R&D on MLOps! From zero to hero in two lines of code
01h 04mIntroduction to ML.NET
09mMachine Learning: Catalyst for Cloud-Native Digital Transformation
50mA Pragmatic Deep Dive into ML.NET
30mMLOps for Continuous Learning
41mMLOps in 30 Minutes
47mMachine Learning in Production: An Intro to MLOps.
25mTinyML as-a-Service