From the course: Deep Learning: Model Optimization and Tuning
Join today to access over 24,700 courses taught by industry experts.
The deep learning tuning process
From the course: Deep Learning: Model Optimization and Tuning
The deep learning tuning process
- [Instructor] Tuning a model should be executed through an organized process making sure that experiments are tracked and the results baselined. Ad hoc tuning will only lead to ineffective results in the long term. We first need to get ready for tuning. For this, the first step is to set clear goals on the outcomes. What are our accuracy or efficiency targets? Are those reasonable? We then need to select the training data and prepare it for experimentation. It is important to choose datasets that are balanced across various classes and cover a wide range of real-world samples. We should plan for testing and validation of models with independent data. Multiple real-life use cases should be covered and production-like scenarios should be used for measuring model performance. What are the key levers available in a deep learning model that can be experimented with? We have the network architecture levers like layers,…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.
