In this course, you’ll learn the basics of deep learning, and build your own deep neural networks using PyTorch. You’ll get practical experience with PyTorch through coding exercises and projects implementing state-of-the-art AI applications such as style transfer and text generation.
Introduction to Deep Learning
Discover the basic concepts of deep learning such as neural networks and gradient descent,Implement a neural network in NumPy and train it using gradient descent with in-class programming exercises,Build a neural network to predict student admissions
Introduction to PyTorch
Hear from Soumith Chintala, the creator of PyTorch, how the framework came to be, where it’s being used now, and how it’s changing the future of deep learning
Deep Learning with PyTorch
Build your first neural network with PyTorch to classify images of clothing,Work through a set of Jupyter Notebooks to learn the major components of PyTorch,Load a pre-trained neural network to build a state-of-the-art image classifier
Convolutional Neural Networks
Use PyTorch to build Convolutional Neural Networks for state-of-the-art computer vision applications,Train a convolutional network to classify dog breeds from images of dogs
Use a pre-trained convolutional network to create new art by merging the style of one image with the content of another image,Implement the paper "A Neural Algorithm of Artistic Style” by Leon A. Gatys, Alexander S. Ecker, and Matthias Bethge"
Recurrent Neural Networks
Build recurrent neural networks with PyTorch that can learn from sequential data such as natural language,Implement a network that learns from Tolstoy’s Anna Karenina to generate new text based on the novel
Natural Language Classification
Use PyTorch to implement a recurrent neural network that can classify text,Use your network to predict the sentiment of movie reviews
Deploying with PyTorch
Soumith Chintala teaches you how to deploy deep learning models with PyTorch,Build a chatbot and compile the network for deployment in a production environment
Luis Serrano, Alexis Cook, Soumith Chintala, Cezanne Camacho and Mat Leonard
Dimitrios Tosidis completed this course, spending 2 hours a week on it and found the course difficulty to be medium.
I took the challenge from Facebook. I wish I could have more time, I spent around 23 hours, although I had almost 10 weeks for it. The course is very interesting, as they give a lot of code to experiment with. The several methods need a lot of time and computation power, you have to use a public cloud for AI. The course has the scope of image recognition, but it goes further with text recognition with methods that are developed during the last 4 years, such as convolution neural networks and recurrent neural networks.
Anonymous completed this course.
I completed this course because I waned to apply a Bertelsmann scholarship on Udacity, it's a great course with a pretty useful stuff on PyTorch, very practical with a complete repo on GitHub , 4/5