When it comes to doing something serious, I believe in systems. Like the author Scott Adams says in his book How to fail at almost everything and still win big

I’ve been working on wrapping up my research projects and writing manuscripts these days (as a 6-th year grad student, clock is ticking). Last week, the writing center in our university offered a visual 2-day dissertation retreat. I hesitated whether I should join when I saw the advertisement last month, because “they will talk about things everyone knows”, “their writing schedule doesn’t work for me”, “my simulations are not finalized, I can’t write like that”. But part of me is saying “These people are professional at writing, they can teach me something.” So I signed up. It happened on Thursday and Friday. It turned out to be very helpful and I wanted to share what I learned below!

Everyone is publishing something these days: either about their life or their work.

I came across this very practical online course Introduction to Machine Learning for Coders(http://course18.fast.ai/ml) on Twitter. I watched the first 7 lessons and enjoyed it so far. I was able to see how Jeremy approach these problems and got many tips and tricks along the way.

object oriented programming

I’ve been working on a method research project lately. It’s basically a modified version of a current available algorithm. But we modified it to make it suitable to some new conditions the original method not designed to handle.

Today I watched this lecture by Professor Patrick Winston lecture video.

I came across this book How to measure anything: finding the value of “intangibles” in business on Twitter. Someone highly recommended the book.

新年的第一天，特别想在上面写写去年感受最深的事，2019年5月以后，可能是因为不用上课了，有了很多自己的时间（几乎全是自己的时间。。。），因此也要非常直接的面对专属于自己的，没有固定解法的问题。以前读书，做题，考试那一套经验变得很不实用，于是需要学学新的思维方式和工作方式，虽然做的还不好，收获还是挺大的

Lately, I’ve seen these words a lot: “big data ecosystem”, “hadoop”, “spark”, “mapreduce”, “hive”, “pig” etc.

I’ve been having this question for so long: is there some ways I can save my own code templates?

We often predict the probability of occurence of binary outcomes: e.g. cases or controls using logistic regressions.

Although there are many nonlinear “fancy” models, we know linear models still have advantages in inference and works well in real world.

At work, sometimes we need to combine and analysis information from lots of output files together.

My advisor used to tell me: read broadly to gain knowledge; read with questions in mind; grasp the main message from papers.

In our work, we often need to generate lots of similar commands at once. There are lots of ways to do this, I just want to write about the one I used the most often: Parallel

In statistical genetics field, we often observe non-normal distributed data. Also, we use linear models and parametric tests a lot. Linear models and parametric tests often assume residuals are normaly distributed. Then how do we respond to non-normality?

Hi there, this is my first post. Currently, I am a 4th year graduate student. Maybe I should know how to be a graduate student very well by now. But the truth is that I am still constantly facing new situations that I am not sure what to do. I want to write about my recent struggles on accomplishing research tasks and resource I found helpful to me when dealing with these problems.

Here is some bold text