1. what cost look like? W=2 의 값일때 cost(W)의 값은 4.67로 동일하다 Y축 : Cost X축 : W로 두었을 때, minimize 시킬 수 있어야한다. 2. Gradient descent algorigm (1) minimize cost function (2) Gradient descent is used minimization problems (3) for a given cost function 3. how it works?gradient descent the lowest point부분을 찾는다.(경사가 있는 그래프의 minimize) 경사를 어떻게 구할 수 있을까? 미분을 사용한다.미분을 사용한다는것은 ? 그 그래프의 기울기를 구한다고 생각하면 쉽다. Ex)미분의 예제 C..
1. Hypothesis and cost functioncostfunction의 기본 원리 및 구조 2. 이제 tensorflow 실습-1 import tensorflow as tftf.set_random_seed(777) # for reproducibility # X and Y data 학습데이터 X,Yx_train = [1, 2, 3]y_train = [1, 2, 3] # Try to find values for W and b to compute y_data = x_data * W + b# We know that W should be 1 and b should be 0# But let TensorFlow figure it out Variable은 변수와 다른 개념인데, 텐서플로우가 자체적으로 변경시키..
1. Linear Regression Predicting exam score : regression supervised learning 을 통해 알아보자.supervised learning은 간단히 데이터를 가지고 학습을 시키는것이다. learning course : Train -> regression 간단히 말하면,데이터를 trainning 시켜놓은 후, 그 조건에 맞는 데이터에 대한 확률을 regression analysisX,Y의 표로 시행되고 있음을 알면된다. (1) Linear Regression이란?(hypothesis)데이터 분석에 있어서 하나의 가설 검증이 필요하다. 이 가설을 검증할때, Linear 한 모델이 우리 데이터에 맞을것이다. 라는의미 가설을 세우고, 가장 적합한 데이터를 결정하..
앞서 강좌에서 Tensorflow의 환경 구축을 완료해보았습니다. 1.기본 소스코드 분석>>> import tensorflow as tf >>> hello = tf.constant('Hello, TensorFlow!') >>> sess = tf.Session() 2018-03-11 00:55:58.218567: I tensorflow/core/platform/cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA>>> print(sess.run(hello)) Hello, TensorFlow! 이 코드는 앞선 강좌에서 환경구축을 완료한후 기본 소스코드..
1. What is Tensorflow? - tensorflow is computer using data flow graph. - python(library) 2. what is a data flow graph?돌아다니는 데이터 = tensor흐름 = flow tensor + flow = tensorflow 3. Mac Tensorflow Env Setting(virtualenv 사용하여 Tensorflow 환경 구축하기) 가상으로 텐서플로우 환경을 만들어서 실행시킬것이다. 그 이유는, 기존의 시스템의 파이썬과의 충돌을 방지하기 위함(Anaconda도 많이 사용하는데 공식적으로 지침은 아니라고 한다.)필자는 virtualenv를 사용하여 설치하였다. (1) 터미널을 켠다(Mac기준)단, 여기에서는 파이썬..
- ML Algorigm 종류1. Linear regression2. Logistic regression(classfication)3. nerual Network - tool : tensorflow- note 참고 자료 : http://www.holehouse.org/mlclass/ 1. what is ML?* Spam Filter : many rules* Automatic Driving : too many rules 2. Supervised / Unsupervised ?(1) Supervised 와 Unsupervised를 나누는 기준은 학습하는 방법에 따른 분류에 따라 바뀐다. (2) Supervised learning에서는 카테고리가 있는 것 예를 들어 , 고양이,강아지등 동물을 카테고리 별로 나누..
- Total
- Today
- Yesterday
- 감자코딩
- 백준알고리즘
- Spring
- programming
- 안드로이드
- 코드엔진
- BFS
- Android
- 텐서플로우
- 알고리즘
- 노드
- db
- Algorigm
- 감자개발자
- 백준
- C langauge
- TensorFlow
- node.js
- node
- 리버싱
- 학교
- MVC
- 프로그래밍
- 복습
- 개발하는 관광이
- 스프링
- 초보자를 위한 C언어 300제
- Controller
- 머신러닝
- C언어
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 |