ML Study

Goal

1. Study ML/DL/DS.
2. Do Projects to check our achievement.
3. Challenge Contests/Competition such as Kaggle
4. Study CS and Coding Test.

Participants

	YoungJ-Baek	jeonjw25	HoneyHon
Role	Study Group Leader	Github Master Project Leader	Slack Master Project Leader
Job	Embedded/Firmware Engineer	HUFS student(Senior)	bachelor's degree(KHU)
Major	Electrical Engineering	Information Communication Engineering	Industrial Engineering
Interest	Image Processing/ML/DL ML in Semiconductor	Computer Vision/ DL Drone Control	Optimization
Link	Github	Github	Github

INDEX

1. (09.11) Machine Learning Introduction
2. (09.18) Perceptron(Neuron)
3. (09.25) Gradient Descent
4. (10.02) Logistic Regression
5. (10.09) Support Vector Machine(SVM)
6. (10.16) Decision Tree & Random Forest
7. (10.23) K-Nearest Neighbor(KNN)
8. (10.30) Normalization & Standardization & Regularization
9. (11.06) Feature Selection
10. (11.13) Dimensionality Reduction
11. (11.20) Hyper Parameter Tuning
12. (11.27) Ensemble
13. (12.04) Unsupervised Learning(Cluster)
14. (12.11) CNN
15. (12.18) RNN
16. (12.25) GAN
17. (01.01) Reinforcement Learning

TOOL

• Github
• Slack
• Colab for GPU
• C++ & Python for Projects

PLAN

Offline Meeting

1. Paper Review ( 1 person / 1 weak )

   YoungJ-Baek	jeonjw25	HoneyHon
   FlowNet: Learning Optical Flow with Convolutional Networks & FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks (CVPR 2017) @(09.11)	YOLACT: Real-time Instance Segmentation @(09.18)	Swin Transformer: Hierarchical Vision Transformer using Shifted Windows @(10.02)	ImageNet Classification with Deep Convolutional Neural Networks(AlexNet) - 2012 @(09.25)
   Rethinking the Inception Architecture for Computer Vision (Inception-v2~3) - 2016	yolov5	Cascade R-CNN: Delving into High Quality Object Detection	Deep Residual Learning for Image Recognition (ResNet) - 2015
   Densely Connected Convolutional Networks (Dense****Net) - 2017	yolop		EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks(EfficientNet) - 2019
   	complex yolo		Combination of Multiple Global Descriptors for Image Retrieval(CGD) - 2020

2. Code Review

   1. LeetCode
   2. Baekjoon
   3. Programmers

3. Project

   1. Translating Sign Language
      • Master: HoneyHon
      • Status: The literature review
      • Link: Github
      • Reference : Swin Transformer
   2. Drone
      • Master: jeonjw25
      • Status: -
      • Link:
   3. Undefined

4. ML Study

   1. Theory Study
      • Master: YoungJ-Baek
      • Goal: Study basic ML/DL/DS knowledge, and DIY with Opensource Datasets
   2. Challenge
      • Master: jeonjw25
      • Goal: Challenge with out knowledge by Kaggle & Competitions

Online Meeting

1. Github
   • jeonjw25
     • Github Master
     • Project Management
   • HoneyHon
     • Share personal profile editting tips.
2. Slack
   • HoneyHon
     • Daily Check
     • Project Issue
     • Process