Welcome to the Zero to Mastery Learn PyTorch for Deep Learning course, the second most incredible resource for delving into PyTorch (with the PyTorch documentation taking the lead).
- ๐ Online book format: Dive into the course materials presented in an easily readable online book at learnpytorch.io.
- ๐ฅ Initial five sections on YouTube: Get a rapid grasp of PyTorch by watching the first 25 hours of content.
- ๐ฌ Course emphasis: Immerse yourself in code, conduct experiments, and reinforce your understanding through hands-on experience.
- ๐โโ๏ธ Teaching approach: Follow the principles outlined in https://sive.rs/kimo.
Section | Content Covered | Exercises & Additional Resources | Slides |
---|---|---|---|
00 - PyTorch Fundamentals | Numerous foundational PyTorch operations crucial for deep learning and neural networks. | Access exercises & additional resources | View slides |
You: A novice in machine learning or deep learning eager to acquire PyTorch skills.
This course: Guides you through PyTorch and essential machine learning concepts in a practical, code-centric manner.
Even if you possess over a year of experience in machine learning, this course, designed with beginners in mind, can still provide valuable insights.
- 3-6 months of Python coding experience.
- Completion of at least one beginner-level machine learning course (though additional resources are linked for various topics).
- Familiarity with Jupyter Notebooks or Google Colab (you can pick this up as you progress).
- An open-minded willingness to learn (the most crucial requirement).
For items 1 & 2, consider the Zero to Mastery Data Science and Machine Learning Bootcamp, covering fundamental machine learning and Python concepts (full disclosure: I teach that course too).
Access all course materials for free in an online book at learnpytorch.io. If you prefer reading, explore the resources there.
For those inclined toward video learning, the course adopts an apprenticeship-style format, where I write PyTorch code, and you follow suit.
The course advocates if in doubt, run the code and experiment, experiment, experiment! as mantras, emphasizing hands-on learning.
All code is implemented using Google Colab Notebooks (Jupyter Notebooks are also suitable), offering a fantastic free environment for experimenting with machine learning.
Certificates are available for completing the video segments, but the real value lies in the practical experience gained.
Consider this course as a springboard for your machine learning journey. By the end, you'll have crafted extensive PyTorch code, exposed yourself to vital machine learning concepts, and built a foundation for tackling your projects or dissecting public PyTorch-based machine learning endeavors.
Commencing with the essential PyTorch and machine learning fundamentals, the course progresses to cover advanced topics such as PyTorch neural network classification, workflows, computer vision, custom datasets, experiment tracking, model deployment, and the potent technique of transfer learning.
Throughout the journey, you'll undertake three significant projects centered around FoodVision, a comprehensive neural network computer vision model designed to classify images of food. These projects serve as milestones, reinforcing key PyTorch concepts and creating a portfolio to showcase your skills to potential employers.
While you can access the materials on any device, a desktop browser is recommended for an optimal viewing and coding experience.
The course utilizes Google Colab, a free tool. If you're unfamiliar, complete the free Introduction to Google Colab tutorial before proceeding.
To commence:
- Click on one of the notebook or section links above, like "00. PyTorch Fundamentals".
- Click the "Open in Colab" button at the top.
- Press SHIFT+Enter a few times and observe the magic unfold.