Deep Learning and PyTorch Bootcamp for Beginners (Easy! From Basics to ChatGPT Core Transformer) [Data Analysis/Science Part3]

For beginners in Python Deep Learning
A high-quality lecture that teaches step-by-step
This is a lecture created by Dave Lee of Jjanjaemi Coding.

Complex AI Technology: Where to Start?

• In recent years, you can learn artificial intelligence technology by learning deep learning technology.
• Unlike other technologies, deep learning technology cannot be implemented immediately and requires an understanding of the theory. You can think of the theory as 80%.
• The problem is that it is difficult to understand the theory all at once, and some parts require knowledge of mathematics, statistics, and probability .
• However, if you only understand the theory superficially, you will not be able to develop the thinking skills to understand the latest deep learning.
• This course covers the essential knowledge and theories needed to learn deep learning at a beginner level.
• Additionally, the implementation is structured so that you can learn various examples and grammar step by step on how to use PyTorch.

The instructor has completely recorded what he felt and worried about after failing numerous times .

• Deep learning is actually difficult to learn!
  • The basic theory is connected to mathematics, statistics, probability, and machine learning technology, and the volume is considerable.
  • Usually, in deep learning lectures, you don't implement the basic deep learning code until the very end.
• However, if you take the basic theory part too thinly, it will be difficult to acquire the fundamentals of deep learning.
• So, this lecture covers the theory in depth enough to learn at an introductory level, organizing the necessary related knowledge.
• It is structured so that you can learn the theory and implementation in parallel without getting tired.

👉 Ultimately, if you listen to the lecture to the end, you will naturally feel that 'now I have also built up the basics of deep learning.'

Organize the theories that need to be organized one by one ,
From installing PyTorch, we also improved the deep learning code one by one .
In the end, I set it up so that I could submit a Kaggle problem.

💬 When I tried to learn deep learning technology, there were so many things to organize!

That's right. Since deep learning theory is connected to mathematics, statistics, probability, and machine learning, there are too many parts to organize even if you learn one. It takes a considerable amount of time just to find and organize them. This lecture is a lecture that organizes as much as possible to a level that can be understood when learning deep learning for the first time. Like the existing lectures of Janjaemi Coding, we will organize and explain it step by step in Janjaemi Coding's own style.

This alone can save you a lot of time! It goes from beginner level to deep learning!

💬 This is my first time with deep learning! What are the skills I need to learn first to take this course?

Basically, anyone with light experience with Python, pandas, data visualization (plotly), and machine learning libraries (sklearn) will be sufficient. All related background knowledge, including mathematics required to understand deep learning, is covered in this lecture.

If you lack the above skills, we recommend taking this course along with the following lectures.

A good lecture to listen to together

First, through the beginner's Python data analysis (data part 1) course, you will learn Python, pandas, data visualization (plotly), and basic exploratory data analysis techniques. After that, you need to become familiar with learning-related processes, basic mathematics, probability, and statistics through the beginner's Python machine learning boot camp course . If you learn deep learning technology based on this, you can learn deep learning theory and ChatGPT's core technology more quickly .

💬 I am a beginner considering a career in data. How can I learn systematically?

It will be helpful if you take the data analysis/science course shown just above. Data-related careers can be broadly divided into data analysts and recent data scientists. In the end, both careers require that you can collect, store, analyze, and predict data through programming. In addition, if you build knowledge of each business field (called domain knowledge), you can be competitive. We also provide a data analysis/science roadmap so that you can systematically learn the entire data process in a short period of time for your data career. You can check the roadmap at the bottom of this page.

The Data Analysis/Science Roadmap is designed to help you build a solid foundation in data technology, with a curriculum that has never been done before, and with a level of difficulty in mind. These are proven lectures that many people have studied over the years and have given very good feedback.

💬 How difficult is it to learn deep learning technology?

It is true that it is more difficult than you think. However, if you organize it step by step, it is a technique that you can eventually make your own.

The most difficult part when learning deep learning for the first time is studying mathematics, statistics, and probability to understand the related theories . Even if an instructor who has studied related technologies for decades explains it easily, it takes a long time for someone to learn it.

If you make a mistake in one of these, there is no end. You need to adjust the pace. You can learn the parts that you can understand step by step, and then move on to the next step. This lecture has been organized to a level that even beginners in deep learning can understand, considering this adjustment of pace. Wise people focus on the parts that they need to focus on at this stage.

💬 Recently, there have been many Kaggle competitions that solve real data problems. Is it possible?

This course covers various implementation techniques and examples, and explains them step by step so that you can submit actual Kaggle problems.

• First, let's start with the theory and PyTorch syntax.
• Step by step, we will move forward with slightly improved code and examples.
• Finally, we explained the steps to apply it to Kaggle problems.

This lecture serves as an introduction for those who are learning deep learning for the first time .

With the thought that I was learning for the first time,
So that even beginners can acquire the basics of deep learning in a short period of time!

Ah, I can do deep learning too! I feel really happy when I feel that way. I can understand and utilize the pinnacle of knowledge created by mankind, deep learning! This feeling soon turns into pride. Try cutting-edge new technologies as much as you can! Even if you only look at the big picture, it is clearly different.

💾 Increase your learning effectiveness with easily understandable summarized materials and codes!

There is an abundance of materials and information. After listening to a lecture that explains in detail the summary material that allows you to understand only the essential parts, you can immediately understand it by looking at the material whenever you think, "Oh! There was this kind of content?"

It contains only the essential parts needed to help you understand the topic in a concise manner.

• We provide deep learning implementation code files. Test code is provided in a format (Jupyter notebook format) that allows for code testing, and basic theories are provided as PDF files.
• We provide PDF materials related to deep learning so that you can check them at any time like an ebook. (However, copying and downloading of related materials is restricted due to copyright issues.)