일 | 월 | 화 | 수 | 목 | 금 | 토 |
---|---|---|---|---|---|---|
1 | ||||||
2 | 3 | 4 | 5 | 6 | 7 | 8 |
9 | 10 | 11 | 12 | 13 | 14 | 15 |
16 | 17 | 18 | 19 | 20 | 21 | 22 |
23 | 24 | 25 | 26 | 27 | 28 | 29 |
30 | 31 |
- kotlin
- JanusWebRTCServer
- pytest
- 티스토리챌린지
- 겨울 부산
- VARCHAR (1)
- table not found
- mp4fpsmod
- k8s #kubernetes #쿠버네티스
- 달인막창
- 오블완
- Spring Batch
- 헥사고날아키텍처 #육각형아키텍처 #유스케이스
- python
- tolerated
- taint
- Value too long for column
- JanusWebRTC
- 코루틴 컨텍스트
- JanusWebRTCGateway
- 개성국밥
- 깡돼후
- 코루틴 빌더
- JanusGateway
- PytestPluginManager
- terminal
- PersistenceContext
- preemption #
- 자원부족
- vfr video
목록Data Analysis/Machine learning (52)
너와 나의 스토리
jupyter notebook에서 코드 작동 시키는 법 1. anaconda prompt켜서 다음 명령어 치기 activate tensorflow35

강의: www.coursera.org Lecture 4 Linear Regression With Multiple Variables Multiple feature 함수 n=feature개수 다중 변수인 경우 Xj의 각 값이 반영된다. GD: Feature Scaling idea: Make sure features are on a similar scale. x1과 x2의 값 차이가 많이 나면 왼쪽처럼 얇고 긴 타원 모양이되고, gradient descent의 최적값 찾는 경로가 복잡하고 많이 이동하게 된다. 오른쪽 같은 경우는 x1과 x2가 비슷하다. 이처럼 각 변수 값이 일정해지면 , gradient descent의 최적값을 짧은 경로로 쉽게 찾을 수 있다. Feature Scaling -> 모든 feat..

강의: www.coursera.org Lecture 2.1 Model Representation ●기호 표기 - x^i, y^i [x(i), y(i)] 이런 식으로 쓰이는데, 여기서 i는 제곱이 아니라 인덱스이다. x^3이면 x 자료에서 3번째 값을 의미하는 것이다. ●지도 알고리즘 Lecture 2.2 Cost Function ㄴ 선형회귀(Linear Regression)에 대한 전반적인 목적 함수(objective function)이다. * objective function = cost function = 오차함수의 제곱 최소 training set에서, x값이 주어진다면 우리는 y값을 예측할 수 있다. h(x)와 y간의 차이를 작게 만들고 싶을 때 Θ0, Θ1: 최소값 x(i): 예측된 집 가격 y..

강의: www.coursera.org Lecture 1.1 Introduction: Machine Learning Machine Learning: 데이터를 이용해서 컴퓨터를 학습시키는 방법론 Tom이 정의한 머신 러닝: a well-posed learning problem is defined as follows he says “A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E.” Q: Suppose your email progra..