R

The problem I faced I am a R user who love using R Shiny to do any interactive visual related projects. My colleague who is more familiar with Python love to use Dash to do his dashboard tasks. One day, our manager asked us if we two could work together to create an internal dashboard. We looked at each other and started thinking which path is easier, have me learn Dash from the beginning or have him learn R Shiny from the beginning?

In my work, I usually deal with dataset of products from different customers across different market places. Basically, each product has its own time series dataset. The size of each dataset is not big but we have millions of them. Before finding any convincing reasons to combine dataset from different products, now we just treat them all as independent dataset. And, since they are all independent, it is definitely a good idea to use parallel computing to push the limit of our machine and to make code executed efficiently.

When I am working on my personal projects, all the data I have can be easily saved as .csv or .txt files. Things are not complicated. However, when things are in scale, it is hard to not to talk about database. In my company, all the data are stored in SQL Server database. So, it is important for me to know how to connect R with SQL Server. And, also good to know how to interact with database in R by passing variables in R environment to database and then return the desired dataset.

The Github repository for this website : choux130/slide_thesis_ioslides. The link to my slides : https://choux130.github.io/slide_thesis_ioslides/#1 Having an opportunity to give a presentation for my master thesis, I decided to give it a try on R Markdown Presentation with interactive graphs and planned to publish it online after the presentation. There are three main choices in R Studio for the R Markdown Presentation: ioslides, Slidy, and Beamer. Beamer is for .

This post shows the R code for LDA and QDA by using funtion lda() and qda() in package MASS. To show how to use these function, I created a function, bvn(), to generate bivariate normal dataset based on the assumptions and then used lda() and qda() on the generated datasets. Details Resources for Package ‘MASS’ CRAN - Package ‘MASS’ Package ‘MASS’ - Reference manual Example Code LDA : Suppose our dataset are from \[X_{green}=N(\begin{pmatrix} 0.

Update : 2018-05-06 Yihui Xie has announced a new package blogdown in September 2017. This package include all the features in knitr-jekyll but instead of using Jekyll this time he switched to Hugo. There is a terrific instruction from him, blogdown: Creating Websites with R Markdown. So, if you want to integrate your website with R Markdown, I will definitely recommend you to check that out. I also convert my whole website from Jekyll with knitr-jekyll to Hugo with blogdown.