Tian Zheng

Here is why Zheng rocks: 1) The workload was very manageable. She adjusted deadlines when she saw that we were stressed and needed more time. She believes students should learn in a nurturing environment, so she lets you submit homework multiple times (until you master the content). 2) Her final exam was about testing your knowledge, not tricking you. Zheng even offered to increase the weight of the final for individual students who were worried they might not pass the class. This was ingenious-it allowed more students to do well and also created a greater incentive to study for those who felt their grade was a lost cause. 3) There were ample office hours where you could ask questions. Zheng is very approachable and friendly. She stayed VERY late at the office hours before our big coding project was due because she really cares about helping her students. 4) Zheng organized an army of TAs who were always there to answer questions on the course discord she built us. This was really nice for students who worry about asking the professor questions. The TAs also helped us in class when we split into breakout rooms to work on coding notebooks. 5) Zheng added a research credit for students who were interested in exploring coding more on their own. We analyzed election prediction models and proposed additions to them. This was such a great opportunity, especially because it was acces…

Fall 2020, Prof Zheng decided to make Intro to Stats election-themed which was good in theory but terrible in practice. Every Wednesday we had a guest lecturer talk about election stats which had nothing to do with the class. Every Monday we did a review of the stats formulas and assignments that we needed to teach ourselves. We also did labs with R code which was fine because I've coded before but it wasn't really coding it was just hitting buttons and analyzing the graphs that would show up. But since all of our stats knowledge was essentially self-taught attending class was pretty much a waste of time in my opinion. Prof Zheng is nice but she only spoke 25% of the time at most. Overall I can't assess what a normal Intro to Stats would be like because this was far from normal. It was just confusing when stuff was due and if what we were doing in class was even relevant. Tbh I don't think they'll be teaching stats in this format again so you won't have to worry. Other than the confusion, this class was not difficult: weekly assignments and the final were very straightforward and totally open note and open book. Also no midterm!

Tian Zheng is a wonderful professor. She really made this class one of my favorite at Columbia through her humor and teaching style. Applied Data Science is entirely project-based. The course grade is based on five projects - a group of four or five people (randomly chosen) works on each project (except for the last project - you can pick your own partners). Professor Zheng had a great vision for the course, from exploratory analysis of US Census data from Kaggle, to image recognition between cats and dogs. Even though the projects seemed very daunting (especially since most of the students in the course are statistics MA students), Zheng was very helpful and provided useful tutorials and offered valuable guidance on Piazza. It was very clear that she wanted her students to learn and grow, yet she provided enough autonomy for the students to work independently (without lecturing all the time). She even provided snacks in class and got class stickers! On another note, I was initially intimidated by the requirements of the course and how I could stack up against statistics MA students. However, Tian Zheng created a very supportive environment for all of the students and ensured that everyone could use his/her strength to contribute in some way. Because of her, I felt that I learned a lot and made a lot of friends. I highly recommend this class and Professor Tian Zheng. The next…

Amazing! She has a slight accent but is very easy to understand. She's a great teacher, cares about her students, is extremely nice, and is very organized. She really explains concepts well and can help make the sometimes confusing material easy to understand. She also spends time teaching you how to use R. She doesn't test on it, but she makes you use it in the problem sets. A good skill to learn. The material isn't super exciting (unless you're very into statistics) but can be interesting on occasion -- especially with Tian teaching and trying to keep things fun.

I honestly couldn't dislike T.Z. if I tried. She is such a sweet, adorable person. Wicked smart, too -- if you check out her CV. She goes slowly through the material and seems eager to make sure every student understands. Her jokes are ... well, let's just say it's the innocent, untutored delivery that makes people laugh. She's one of the profs you'd like to get to know over coffee.

Climate modeling, machine learning techniques, hybrids which combine physics motivated models and ML techniques. Broad discussion on many topics, channeled through a specific application. Lots of interesting "meta" discussions on research, grant writing, and research proposals.

Tian makes intro stats painless. Her lectures are well-organized, and though they are all Powerpoint-based, she works out a lot of examples on the board. She answers questions during class and always stresses the important points. She also provides in-class copies of lectures and uploads them online afterwards. When she taught us, she didn't hold a lot of office hours, but there were two Tas, each of whom held a recitation during the week (separate from her office hours), so it was easy to get help. Her accent is not an issue, and she's very good at explaining confusing concepts. The final data project is pretty annoying and useless, but the grading is generous, and she provides plenty of resources. In short, this is a very easy class with very fair and transparent expectations. Can be easily tacked on as a sixth class. Tian is understanding of the fact that her class is probably not a priority for most people.

Five projects - each weighed equally. They vary in difficulty - the first one was the easiest while the third one (image recognition) was the hardest, while the students could choose what to do for the fifth one. Make sure to start early and to communicate with your group members. Zheng does require groups to mention whether everyone contributed equally to the project or not.

I've had Tian for two classes: stat 1111 and this course. Over the years her accent has improved immensely, thus making her one of the best statistics professors in the department. Her explanations are straightforward and she includes her own examples, which give students another way to understand the material. The book is very good, straightforward and if you take the prereqs, this class should be pretty easy.

How to behave when dealing with statistical analysis.