Скачать или смотреть #Python

RJ Studio’s 110th video is applying Data Augmentation configuration into Convolutional Neural Network model in Python. ✨
◕ ◡ ◕

In this video, ConvNets are trained to distinguish dogs from cats 🐶😺

First, I train a simple ConvNet (aka CNN) withOUT implementing Data Augmentation.

Next, I show how much improvement Convnet can make once Data Augmentation is used. 🚀


According to the textbook I am reading,
"[Data Augmentation is] a powerful technique for mitigating overfitting in computer vision."


A lot of explanation in this video is from a book titled,
"Deep Learning with Python" (Chapter 5)

Dogs vs. Cats Kaggle dataset is from: https://www.kaggle.com/competitions/d...

Please check my GitHub for code: https://github.com/rjlee13/RJStudio_GH

Python logo in the thumbnail is from https://www.python.org/community/logos/

Kaggle logo in the thumbnail is from https://www.kaggle.com/


☆ Video Chapters ☆
0:00 Intro
1:07 ImageDataGenerator
3:16 CNN Architecture
4:04 Compile & Fit
6:17 Visualize Data Augmentation
7:56 Apply Data Augmentation
9:00 CNN + Data Augmentation
9:36 Compile & Fit
11:07 Outro


2 Background songs in this video are:
"Cheerleader" by Pentatonix
"Wake Me Up" by 2CELLOS
^‿^



Hope my video taught you something new!
And feel free to correct me if I made some mistakes in the comments :)


이번 비디오에서는 합성곱 신경망 모델의 과적합을 방지하기위해 자주 쓰이는 (이미지) 데이터 증강에 대하여 알아봅니다. ≧◡≦
캐글에서 다운로드 받은 데이터로 강아지 vs 고양이를 구별하는 모델을 학습합니다!
유익한 정보이길 바랄게요~ 감사합니다.


“In teaching others, we teach ourselves.”

Instagram:   / musicalum_rj  
LinkedIn:   / rj-lee