codestates / codestates/ds-TIL

[TIL] day81 오세광 201229

Open
#1,277 0 comments 0 reactions 0 assignees View on GitHub
DSFT01
Dominant language
No language data
Stars
2
Forks
1
PR merge metrics
No merged PRs in 30d

Description

# 키워드: convolution, Pooling, 커널, padding, stride, CNN 분류기, 전이학습(Transfer Learning), ResNet50

# (Act)What you actually did.
1. 전체적인 개념을 이해했다.

# (Evaluate)How you evaluate your action's outcome vs. the initial goal(plan).
1. Convolution은 필터의 개념의 새로운 함수를 생성하여 연산하고 다음 레이어로 전달하는 가중치를 새롭게 만들어 줍니다.
2. Cnn학습과정에서 커널은 학습된다
3. Pooling으로 차원을 축소한다.
4. Padding 데이터 사이즈 보존, 데이터를 골고루 쓴다.
5. Stride 얼마만큼씩 움직이며, 이미지를 볼지 결정하는 수
6. 전이학습 기존의 데이터로 학습된 네트워크를 재사용 가능하도록 하는 라이브러리

# (Reflect)
How you worked throughout the stages plan-research-act-observe-evaluate.
What did you do well, what can you improve upon.
How can you improve your process?

1. 앞으로 실제 코드로 구현하는 걸 해보자!
2. 데이터 셋 불러오는 것과, 전이 학습, 그 위에 CNN분류기 만드는 걸 잘 구현해보자

Contributor guide

No contributing guide indexed for this repository

Research direction

Start with the issue body, which records concepts including convolution, pooling, padding, stride, CNN classification, transfer learning, and ResNet50. No source files, tests, or implementation entry points are mentioned; the stated next step is to implement dataset loading, transfer learning, and a CNN classifier, but completion criteria are not defined.

Written by the indexing model from the issue text.

Assessment

Tech stack
machine-learning
Domain
documentation, machine-learning
Issue type
Documentation
Difficulty
1/5
Estimated time
Under an hour
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
15/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.