[机器学习] --- Getting Started With MachineLearning
一. What’s machine learning
Machine Learning is the science of gettingcomputers to act without being explicitly programmed --- Andrew Ng
Machine learning is a technique of data science that helps computers learn from existing data in order to forecast future behaviors, outcomes, and trends. --- Microsoft
二. The difference between traditional approach and Machine Learning
From business problem to Machine learning problem: a Recipe
Step-by-step “recipe” for qualifying a business problem as a machine learning problem
- Do you need machine learning?
- Can you formulate your problem clearly?
- Do you have sufficient examples?
- Does your problem have a regular pattern?
- Can you find meaningful representations of your data?
- How do you define success?
三. How to create machine learning models
四. The Core --- Data
Data understanding
Data Preparation
Approaches for Feature Selection
五. Modelling
Train the model
六. Model Evaluation
Hold-out validation strategy
k-fold cross validation strategy
leave-one-out cross validation strategy
Because future instances have unknown target values, you need to check the accuracy metric of the ML model on data for which you already know the target answer, and use this assessment as a proxy for predictive accuracy on future data 1.
Evaluate your trained model by using validation/test dataset. You compare the results of your model's predictions to the target values in the evaluation data and use statistical techniques appropriate to your model to gauge your success.
What’s the accuracy
Accuracy measures the ratio of correct predictions to the total number of cases evaluated
Increasing precision reduces recall, and vice versa. This is called the precision/recall tradeoff
- Within any one model, you can decide to emphasize either precision or recall.
- You can influence precision and recall by changing the threshold of the model.
Metrics for evaluating regression model
Summary: Testing and Error Metrics
Tuning the Hyperparameter
Model Deployment
[机器学习] --- Getting Started With MachineLearning相关推荐
- 【机器学习】数值分析02——任意方程求根
任意方程求根 全文目录 (博客园)机器学习 (Github)MachineLearning Math 1.简介 方程和函数是代数数学中最为重要的内容之一,从初中直到大学,我们都在研究着方程与函数,甚至 ...
- 【机器学习】数值分析01——绪论及误差分析
数值分析--绪论及误差分析 文章目录 数值分析--绪论及误差分析 全文目录 数值分析的作用及其学习工具使用 数值分析常用工具 数值分析的具体实例(多项式简化求值) 计算机数值误差产生机理 计算机的数值 ...
- 谷歌云端语音识别软件
谷歌云端语音识别软件是是一款非常厉害专业的语音识别神器,粤语也可以直接专为文字,而且准确度极高,它内置强大的语音识别功能,含有全球常用的八十多种语言,高超的智能识别引擎,几乎零误差,多格式支持,速度惊 ...
- 大数据工程师修炼笔记
想要成为大数据工程师?你需要掌握以下知识(上) 文| 林肯公园(拒绝任何不标明来源的转载) 国际数据公司IDC预测,到2020年,企业基于大数据计算分析平台的支出将突破5000亿美元,大数据解决方案在 ...
- 数据驱动的营销方式和加拿大禁止Clearview AI
人工智能 (ARTIFICIAL INTELLIGENCE) The debate on facial recognition, surveillance, and privacy continues ...
- PIE-engine APP教程 ——基于水体指数或监督分类方法的水体频率计算
本次我们来查看进行水体处理的分布,我们这里首先对数据进行预处理,先进行NDWI,AWEI.MNDWI等计算函数和去云函数,第二部分市机器学习部分这里有三个机器学习模型,分别是随机森林.贝叶斯和支持向量 ...
- IF: 8+ 基于单细胞 RNA-seq 构建非小细胞肺癌免疫反应的中性粒细胞预后模型
. 单细胞生信分析教程 桓峰基因公众号推出单细胞生信分析教程并配有视频在线教程,目前整理出来的相关教程目录如下: Topic 6. 克隆进化之 Canopy Topic 7. 克隆进化之 Cardel ...
- 【学习笔记】李宏毅-机器学习MachineLearning
1.机器学习,就是找一个函式 MachineLearning ≈ Looking for Function 2.机器学习两大类任务: Regression:The function outputs a ...
- MachineLearning 13. 机器学习之降维方法UMAP及可视化 (umap)
点击关注,桓峰基因 桓峰基因公众号推出机器学习应用于临床预测的方法,跟着教程轻松学习,每个文本教程配有视频教程大家都可以自由免费学习,目前已有的机器学习教程整理出来如下: MachineLearnin ...
最新文章
- 北大吴思教授:人脑视觉机制有望开启下一代深度神经网络
- [LeetCode]题解(python):087-Scramble String
- Power BI 可视化交互/视觉对象交互
- 欠阿里云一分钱,会是什么样的后果。。。
- C# WebProxy POST 或者 GET
- django ORM相关的那些操作汇总
- Exceptions
- mysql xtrabackup-v2_pxc wsrep_sst_method均配置为xtrabackup-v2报错
- UIWebView UITextView
- WebService 出现因 URL 意外地以“/HelloWorld”结束,请求格式无法识别。
- hdu 敌兵布阵(线段树之单点更新)
- 验证堆表(heap table)存储方式
- LintCode Python 简单级题目 112.删除链表中的重复元素
- 操作简单、功能务实——四维星软件
- ARM上移植ser2net源码
- fins协议握手信号服务器响应,欧姆龙OMRON PLC之HostLink通讯协议(四)-FINS命令工作模式篇...
- 增大图像感受野方法的总结
- 使用腾讯云轻量应用服务器搭建dplayer播放器弹幕服务器
- 如何通过命令提示符进入MySQL服务器
- 数仓工具—Hive实战之UDF汉字首字母(22)