0% completed

All LessonsFree Lessons (5)

Learn Deep Learning

About this Course Why Learn Deep Learning?Overview of the Course

Neural Networks

Linear Classifiers Optimization and Gradient Descent Neural Networks Backpropagation Algorithm Build a Neural Network With Pytorch Quiz Yourself on Neural Networks

Training Neural Networks

Optimization Popular Optimization Algorithms Activation Functions Training in Pytorch Quiz Yourself on Training Neural Networks

Convolutional Neural Networks

The Principles of the Convolution

Convolution in Practice

Build a Convolutional Network

Batch Normalization and Dropout

Skip Connections

CNN Architectures

Quiz Yourself on CNNs

Recurrent Neural Networks

A Simple RNN Cell LSTM: Long Short Term Memory Cells Writing a Custom LSTM Cell in Pytorch Connect LSTM Cells Across Time and Space Quiz Yourself on RNNs

Autoencoders

Generative Learning Basics of Autoencoders Variational Autoencoder: Theory Variational Autoencoder: Practice Quiz Yourself on Autoencoders

Generative Adversarial Networks

Generator and Discriminator Generative Adversarial Networks in Detail Develop a GAN with Pytorch Quiz Yourself on GANs

Attention and Transformers

Sequence to Sequence Models Attention Key Concepts of Transformers Self-Attention Multi-Head Self-Attention Transformers Building Blocks The Transformer's Encoder The Transformer's Decoder Build a Transformer Encoder Quiz Yourself on Transformers

Graph Neural Networks

Basics of Graphs Mathematics for Graphs Graph Convolutional Networks Implementation of a GCN Quiz Yourself on GNNs

Conclusion

Where to go from here?Appendix

Course Assessment

Final Quiz

Introduction to Deep Learning & Neural Networks/

...

/

Build a Convolutional Network

Build a Convolutional Network

Develop a Convolutional Neural Network (CNN) with Pytorch.

We'll cover the following...

It is time to apply what we learned about CNNs through an assignment. We will build a fully functional CNN and train it with our familiar CIFAR10 dataset.

The CNN structure will be:

A conv layer with 3 channels as input, 6 channels as output, and

...