Join our 4-night Hands on Deep Learning: From Fundamentals to State of the Art with "Deep Learning with Python Book."
Class will be held on June 12, June 14, June 19 and June 21 from 6-9pm. In addition to getting 12 hours of instructor led curriculum, you will also receive a copy of "Deep Learning with Python."
Main Topics:
During this course on deep learning, you will learn how to build neural networks from scratch. We will discuss the math involved, which includes calculus, linear algebra, and probability theory, so you'll be able to understand how your model works at a fundamental level.
You will start by building a single-neuron network and gradually build larger and more complex models able to predict more complex phenomenon. We will then learn about Keras and TensorFlow and use those tools to build out our deep learning models rapidly.
We will be building fully connected networks (multilayered perceptrons), convolutional neural networks (CNN) and recurrent neural networks (RNN). In addition, we will be leveraging Transfer Learning along with models like ResNet or Inception, to help boost the predictive power of the network. These models will allow us to predict continuous values or classify images, text or audio.
During the course, you'll learn about how to build a deep learning pipeline, which includes:
Acquiring data
Cleaning data
Normalizing data
Splitting data into test, train, and validation
Model training
Epochs and Batches
Loss functions and Gradient Descent
Optimizers
Validation metrics
You will be shown how to train your models locally and in the cloud using Amazon Web Services' Tesla GPUs.
Overall, at the conclusion of the course, you will have a firm understanding of neural networks and their design, with the ability to design and build your own.
Prerequisites:
- Bring a Laptop (Either a Mac, Linux or a Windows machine running Linux in a VM)- Generally comfortable with mathematics
Setup:
• Bring your laptop with latest Python installed and power cable.
Questions:
If you have questions, or simply want to make sure that this workshop is for you, please reach out to us at
[email protected].