CTD - Python 200 - Lesson4 ML-Deep Learning

Topics Introduction to neural networks and deep learning A brief introduction to neural networks and deep learning. What are they, how do they work, how do they learn? Introduction to pytorch An overview of the PyTorch library, mainly focusing on tensor operations (in the world of deep learning, they use the word "tensor" instead of "array", but it's still just arrays of numbers like we saw with NumPy). Training your first neural network Before diving into more complex models, we will start by training a simple neural network in PyTorch to gain an understanding of the basic workflow for training and evaluating a model -- a workflow that carries over to much more complex architectures. Machine vision intro For this section, we will use a pre-trained convolutional neural network (CNN) to classify images. We use this to show how deep neural networks can be used for computer vision tasks. Transfer learning You rarely train a complex neural network from scratch. Here, we will explore transfer learning, in which a neural network trained on one task is fine-tuned on a similar task, but requires much less data. Many deep-learning pipelines are really just transfer learning pipelines under the hood.