Fine-Tuning Machine Learning Models with DVC Experiments for Transfer Learning

Too Long; Didn't Read

There are plenty of machine learning models available that have been trained to solve one problem and the knowledge gained from that can be applied to a new, yet related problem. For example, a model like AlexNet has been trained on millions of images so you could potentially use this to classify cars, animals, or even people. In this post, we'll be fine-tuning the AlexNet model and the SqueezeNet model to classify bees and ants. We start by initializing these models so we can get the number of model features and the input size we need. We'll use DVC to handle experiments for us and we'll compare the results.

---