4.34 out of 5
4.34
1033 reviews on Udemy

Practical Deep Learning with PyTorch

Accelerate your deep learning with PyTorch covering all the fundamentals of deep learning with a python-first framework.
Instructor:
Deep Learning Wizard
4,197 students enrolled
English [Auto-generated]
Effectively wield PyTorch, a Python-first framework, to build your deep learning projects
Master deep learning concepts and implement them in PyTorch

Growing Importance of Deep Learning

Deep learning underpins a lot of important and increasingly important applications today ranging from facial recognition, to self-driving cars, to medical diagnostics and more.

 

 

 

Made for Anyone

 

Although many courses are very mathematical or too practical in nature, this course strikes a careful balance between the two to provide a solid foundation in deep learning for you to explore further if you are interested in research in the field of deep learning and/or applied deep learning. It is purposefully made for anyone without a strong background in mathematics. And for those with a strong background, it would accelerate your learning in understanding the different models in deep learning.

 

 

 

Code As You Learn

 

This entire course is delivered in a Python Notebook such that you can follow along the videos and replicate the results. You can practice and tweak the models until you truly understand every line of code as we go along. I highly recommend you to type every line of code when you are listening to the videos as this will help a lot in getting used to the syntax.

 

 

Gradual Learning Style

 

The thing about many guides out there is that they lack the transition from the very basics and people often get lost or miss out vital links that are critical in understanding certain models. Because of this, you can see how every single topic is closely linked with one another. In fact, at the beginning of every topic from logistic regression, I take the time to carefully explain how one model is simply a modification from the previous. That is the marvel of deep learning, we can trace back some part of it to linear regression where we will start.

 

 

 

Diagram-Driven Code

 

This course uses more than 100 custom-made diagrams where I took hundreds of hours to carefully create such that you can clearly see the transition from one model to another and understand the models comprehensively. Also, the diagrams are created so you can clearly see the link between the theory that I would teach and the code you would learn.

 

 

 

Mentor Availability

 

When I first started learning, I wished I had a mentor to guide me through the basics till the advanced theories where you can publish research papers and/or implement very complicated projects. And this course provides you with free access to ask any question, no matter how basic. I will be there and try my very best to answer your question. Even if the material is covered here, I will take the effort to point you to where you can learn here and more resources beyond this course.


Math Prerequisite FAQ

This is not a course that emphasizes heavily on the mathematics behind deep learning. It focuses on getting you to understand how everything works first which is very important for you to easily catch up on the mathematics later on. There are mathematics involved but they are limited with the sole aim to enhance your understanding and provide a gentle learning curve for future courses that would dive much deeper into it. 

Latest Python Notebooks Compatible with PyTorch 0.4 and 1.0

There are very small changes from PyTorch 0.3 for this deep learning series where you will find it is extremely easy to transit over! 

Introduction

1
Introduction

I've uploaded all python notebooks in a zip folder, just run them and you're good to go to follow all the lectures. Take note that these notebooks are slightly different from the videos as it's updated to be compatible to PyTorch 0.4 and 1.0! But the differences are very small and easy to change :)


3 small and simple areas that changed for the latest PyTorch (practice on identifying the changes). You don't need to understand these now, but as you go through the videos, you'll start to realize these slight differences that are very easy to change :)

1. Where you use .cuda() it changes to .to(device)

2. Where you use Variable(tensor) it changes to tensor.requires_grad_()

3. Where you use loss.data[0] it changes to loss.item() to get the loss value

Software Requirements

1
CPU Software Requirements
2
CPU Installation of PyTorch
3
PyTorch with GPU on AWS
4
PyTorch with GPU on Linux
5
PyTorch with GPU on MacOSX

PyTorch Fundamentals: Matrices

1
Matrix Basics
2
Seed for Reproducibility
3
Torch to NumPy Bridge
4
NumPy to Torch Bridge
5
GPU and CPU Toggling
6
Basic Mathematical Tensor Operations
7
Summary of Matrices

PyTorch Fundamentals: Variables and Gradients

1
Variables
2
Gradients
3
Summary of Variables and Gradients

Linear Regression with PyTorch

1
Linear Regression Introduction
2
Linear Regression in PyTorch
3
Linear Regression From CPU to GPU in PyTorch

There is the source code attached that is capable of running on the GPU or CPU.

4
Summary of Linear Regression

Logistic Regression with PyTorch

1
Logistic Regression Introduction
2
Linear Regression Problems
3
Logistic Regression In-depth
4
Logistic Regression with PyTorch
5
Logistic Regression From CPU to GPU in PyTorch

There is the source code attached that is capable of running on the GPU or CPU.

6
Summary of Logistic Regression

Feedforward Neural Network with PyTorch

1
Logistic Regression Transition to Feedforward Neural Network
2
Non-linearity
3
Feedforward Neural Network in PyTorch
4
More Feedforward Neural Network Models in PyTorch
5
Feedforward Neural Network From CPU to GPU in PyTorch

There is the source code attached that is capable of running on the GPU or CPU.

6
Summary of Feedforward Neural Network

Convolutional Neural Network (CNN) with PyTorch

1
Feedforward Neural Network Transition to CNN
2
One Convolutional Layer, Input Depth of 1
3
One Convolutional Layer, Input Depth of 3
4
One Convolutional Layer Summary
5
Multiple Convolutional Layers Overview
6
Pooling Layers
7
Padding for Convolutional Layers
8
Output Size Calculation
9
CNN in PyTorch
10
More CNN Models in PyTorch
11
CNN Models Summary
12
Expanding Model's Capacity
13
CNN From CPU to GPU in PyTorch

There is the source code attached that is capable of running on the GPU or CPU.

14
Summary of CNN

Recurrent Neural Networks (RNN)

1
Introduction to RNN
2
RNN in PyTorch
3
More RNN Models in PyTorch
4
RNN From CPU to GPU in PyTorch

There is the source code attached that is capable of running on the GPU or CPU.

5
Summary of RNN

Long Short-Term Memory Networks (LSTM)

1
Introduction to LSTMs
2
LSTM Equations
3
LSTM in PyTorch
4
More LSTM Models in PyTorch
5
LSTM From CPU to GPU in PyTorch

There is the source code attached that is capable of running on the GPU or CPU.

6
Summary of LSTM

What's Next?

1
What's Next?
You can view and review the lecture materials indefinitely, like an on-demand channel.
Definitely! If you have an internet connection, courses on Udemy are available on any device at any time. If you don't have an internet connection, some instructors also let their students download course lectures. That's up to the instructor though, so make sure you get on their good side!
4.3
4.3 out of 5
1033 Ratings

Detailed Rating

Stars 5
416
Stars 4
404
Stars 3
144
Stars 2
48
Stars 1
24
bf7f63338624ffa558a131248395afc3
30-Day Money-Back Guarantee

Includes

6 hours on-demand video
Full lifetime access
Access on mobile and TV
Certificate of Completion