From the course: Machine Learning Foundations: Prototyping with Edge Impulse
Join today to access over 24,900 courses taught by industry experts.
Training the model - Arduino Tutorial
From the course: Machine Learning Foundations: Prototyping with Edge Impulse
Training the model
- [Instructor] After collecting our data and creating our project's Impulse, the next step is training the learning model. Let's go to Edge Impulse to see how this works. We're back on the project dashboard. There's one thing we need to address in our dataset before we start training. To see this, go to the Data acquisition tab. At the top, it shows how much data has been collected. In our case, three minutes and 21 seconds. Then it shows something called the train/test split. Right now, 100% of our data is tagged for training. Usually, you reserve some of your data for testing your model after training. The test data isn't included in training, so it's a good measure of your model's accuracy. A common balance is to have 80% of your data used for training and 20% for testing. Notice there's a warning symbol next to the train/test split. Hovering over that, we see a notification that points out the problem and suggests clicking on it for help rebalancing the dataset. The panel that…
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.