ML
文章平均质量分 89
Walker0319
这个作者很懒,什么都没留下…
展开
-
Machine Learning Week 11 Application Example and Thank you Andrew Ng
1 Photo OCR1.1 Program description and pipeline1.2 Sliding windowstake pedestrian detection as an exampleusing different rectangles to scan the image, step by step-size or stride, say 4 pixelsText detectiongot the result like this1.3 Character原创 2021-09-09 17:40:56 · 80 阅读 · 0 评论 -
Machine Learning Week 10 Learning with Large Dataset
1 Stochastic Gradient Descent1.1 Alternative algorithm to classic (or batch) gradient descentStochastic Gradient Descent is more efficient and scalable to large data sets, say 100,000,000 or more. It is computational expensive to summarise total examples原创 2021-09-08 17:14:50 · 94 阅读 · 0 评论 -
Machine Learning Week 9 Anomaly Detection & Recommender & ex8
1 Anomaly Detection Algorithm1.1 Algorithm1.2 Developing an Anomaly Detection System1.3 Anomaly Detection vs. Supervised Learnig1.4 Choose Features1.5 Multivariate Gaussian Distribution1.6 Anomaly Detection using Multivariate Gaussian Distribution2原创 2021-09-07 17:42:56 · 67 阅读 · 0 评论 -
Machine Learning Week8 Unsupervised Learning and ex7
1 Clustering1.1 Introduction to Unsupervised LearningUNSL uses an unlabeled training set. We don’t have the vector of expected results, and we only have a dataset of features to find structure. UNSL is good for:Market segmentationSocial network analys原创 2021-08-10 15:39:21 · 126 阅读 · 0 评论 -
Machine Learning Week 7 SVM with Kernels and ex6
1 Optiomization ObjectiveSupport Vector Machine(SVM) is another type of supervised ML algorithm.1.1 cost review of LogReg1.h(x) and y2. cost of LogReg1.2 SVMreplace with SVM cost: cost1 and cost0cost1(z) = max(0, k(1+z)), subscirpt 1 stands for y=原创 2021-08-03 15:23:00 · 110 阅读 · 0 评论 -
Machine Learning week6 Advice for applying ML and ex5
1. Deciding what to try firstsome avenues may be:getting more training examplestrying smaller sets of featurestrying additional featurestrying polynomial features, x1x2, x1^3,…increasing or decreasing lambdabut don’t pick it randomly2 first evalu原创 2021-07-30 11:48:45 · 135 阅读 · 0 评论 -
Machine Learning week4&5 and ex3&4
1. Nueral Networks(NN)Performing LinReg with a complex set of data with many features is very unwieldly.100 features quadratic, 5050 resulting new features50*50 image, n=2500, n^2/2=3125000 new featuresso neural networks is applied with many features1原创 2021-07-19 16:30:43 · 53 阅读 · 0 评论 -
Machine Learning week3 logistic regression ex2
Logistic Regression(LogReg) is an approach to classifiction problems, not regression problems. Don’t be confused by the name.1. Binary ClassificationIn linear regression, output vector y is a continuous range of values. But in logistic regression, y={0.原创 2021-07-02 16:41:03 · 63 阅读 · 0 评论 -
Machine Learning week1-2 ex1
1. linear regress with 1 varible (week 1)1.1 Linear regression model with 1 varibleit is hard to write mathematical formula, so I just give upHypothesish(x) = theta0 + theta1 * xcost function: J(theta0, theta1)Goal: minimize Jsimplyfy first, theta0=原创 2021-06-28 20:12:54 · 87 阅读 · 0 评论