인프런 영문 브랜드 로고
인프런 영문 브랜드 로고
AI

/

Deep Learning & Machine Learning

[AI Basics] Understanding CNN for AI Research Engineers

Even though you studied CNN, you still don't understand it? I will briefly explain the basic operation principles of CNN, only the key parts.

(4.8) 5 reviews

378 students

Computer Vision(CV)
Python
PyTorch
CNN
Thumbnail

This course is prepared for Beginners.

What you will learn!

  • Understanding the basic concepts of Convolutional Neural Network (CNN)

  • Learn how convolution operations and filters work

  • Implementing Convolution Operations Using Numpy and Visualizing Results

  • Implementing Convolution Operations and Visualizing Results Using PyTorch

  • Understanding CNN's learning principles and the meaning of input/output channels

I don't really understand CNN.... 😭

Do you understand exactly what convolution operation is?

Let me first show you a simple example of how convolution is used in computer vision.

CNN is Convolution + Neural Network.

If you don't quite understand CNN, first understand what a convolution is 😀

Features of this course

We've included only the essentials for your precious time.

📌 I will explain step by step what a convolution is.

📌 Let's learn how convolution is used in computer vision.

📌 Let's implement Convolution in both Python Numpy and PyTorch and check that the results are the same.

📌 I will explain the learning principles of CNN.

I recommend this to these people

I started studying artificial intelligence (AI).
This course is for beginners and can be taken without any special prior knowledge.

I don't really understand CNN.
If you still don't understand CNN, let's study Convolution first.

I want to understand Convolution
Let's take a look at how convolution operations in PyTorch or TensorFlow are actually performed.

Learn about these things.

What is Convolution?

The operation called convolution itself is not actually that complicated. First, let's understand how exactly this operation works.



Convolution examples in computer vision

Let's learn an example of how convolution is used in computer vision. In fact, you are already using convolution a lot.

Convolution - Numpy implementation

Let's implement Convolution in Numpy. Let's dig into how Convolution works exactly.

Convolution - PyTorch implementation

Let's check that the implementation in Numpy matches the implementation in PyTorch. Then, let's understand CNN more deeply.

Things to note before taking the class

Practice environment

  • You don't have to follow the exercises unconditionally. It's okay to just look at the results.

  • The practical environment is explained based on Windows OS.

  • We use Python, Numpy, and PyTorch.

  • We use Anaconda, VScode, and Jupyter Notebook for environment setup.

    • At the beginning of the lecture, we will explain how to set up the environment.

Recommended for
these people!

Who is this course right for?

  • Anyone who wants to learn the basics of CNN

  • People who say they don't understand CNN even after learning it

  • For those who want to understand CNN properly

Need to know before starting?

  • There is no prerequisite knowledge.

  • (Optional) Basic knowledge of deep learning

  • (Optional) Experience using Numpy and PyTorch

  • (Optional) Understanding of linear algebra and basic mathematical concepts

Hello
This is

624

Students

15

Reviews

5

Answers

4.9

Rating

2

Courses

  • 주요 경력

    • (현) 국내 IT 대기업 AI Research Engineer

    • (전) AI 스타트업 AI Research Engineer

  • AI 연구/개발 이력

    • 다수의 AI 프로젝트 진행 및 AI 프로덕트 출시 경험

       

    • 다수의 AI 연구 및 Top-Tier Conference 논문 게재 경험

    • Generative AI 전문가

  • 기타 이력

    • 국내 학회 인공지능 세션 튜토리얼 강사

    • 국내 대기업 AI 강의 초빙 강사

    • 사내 생성 AI 세미나 강사

       

 

Curriculum

All

15 lectures ∙ (49min)

Course Materials:

Lecture resources
Published: 
Last updated: 

Reviews

Not enough reviews.
Become the author of a review that helps everyone!