Familiar with the basics and ready to apply deep learning with MATLAB®? Get started with the hands-on examples in this ebook. You'll learn three approaches to training neural networks for image classification:
- Training a network from scratch
- Using transfer learning to train an existing network
- Adapting a pretrained network for semantic segmentation
You'll also see two examples showing how deep learning models can be applied to time series or signal data.
Read the ebook and learn how to:
- Create and configure network layers
- Adapt network architectures, including convolutional neural network (CNN), directed acyclic graph (DAG), and long short-term memory (LSTM)
- Select the best training options and algorithms
- Use data augmentation and Bayesian optimization to improve training accuracy
- Incorporate spectrograms for speech recognition