STAT 3250

I found this class to be super practical and relatively low stress. Classes are optional office hours and content is learned from online videos posted. Your grade is completely based on 13 assignments (one each week) that are fully autograder based. My biggest tip for this class would be to find a friend who is good at coding so you can check answers on the questions where the autograder tests are hidden. Holt was approachable and kind when answering questions, though sometimes wouldn’t give a straightforward answer on whether something was correct which was frustrating. Overall, while the assignments sometimes took up a bit of time, they weren’t too bad considering that being in class was optional and office hours were an options if I was feeling stuck. Also, it doesn’t matter much which professor you take it with, as you can attend either professors office hours and assignments are the same.

This class is very easy if you are familiar with Python at all. It is also a reverse classroom so you do your weekly homeworks and ask the TAs/professor questions. Attendance is not required and most people don't show up to class at all. No midterms or projects, just 13 weekly assignments throughout the semester. #tCFF23

I took this class over the summer because it filled up so quickly in the spring. Holt was a nice guy and a decent professor. Always willing to answer questions. The format of the class was to watch a few short lecture videos for homework and then come to class and work on the assignments. There were 12 assignments total and you were allowed to work in groups. The class was not too hard, but I had previous Python experience.

When I took this course, your entire grade was composed of 13 weekly homework assignments you had to turn in (literally nothing else-- no exams or anything like that). The entire course is done in python, so it certainly helps if you have taken CS 1110 or have any other coding experience-- however this isn't necessarily required to do well. Overall, the course is focused on various methods to extract data from large data sets (our final assignment used a data set with over a million observations). The homeworks vary in difficulty, but took me on average about 5-6 hours a week to complete. The best advice I could give for this course is to find a study partner/ group that you can work on homeworks with/ compare answers with because sometimes the questions are a bit unclear as to what output it's asking for and it's fairly easy to make simple mistakes. The homeworks were accompanied by video lectures (usually 30-40 mins in total) that were sometimes helpful, though a lot of the material seemed to repeat itself after the 3rd or 4th week. The only slight issue with this course was the grading scale which is point based, but translated into needing a 95.2% overall to get an A which is a bit high. However, if you start the homeworks early enough and are able to attend office hours and compare answers with a group, you should be fine in this class. Also, this course is helpful for simply giving you the basic tools in order to actually perform data manipulation in python. If you are a stats major I would certainly reccomend this course-- plus Jeff Holt is a great guy who is very approachable.

#tCFspring2021

I would honestly recommend taking this class if you have to for the Stat major. You don't have a final exam or project, so the class grade is entirely composed of weekly assignments graded by the grade scope "autograder." They are pretty long and time consuming, but some are harder than others and you are allowed to collaborate with others as long as you're not directly copying code. I would HIGHLY recommend taking with a friend/collaborating on assignments or going to office hours, because if you check answers w others and/or go to office hours with a TA on questions you're stuck on it isn't super hard to do well.

I have mixed feeling about this class. On one hand, Holt's a nice guy, and the class is useful if you want to do statistics with Python. However, there are also quite a few downsides to this class:

1) Grading. A 95.2% is an A in the class, which I think should be illegal. 13 assignments make up your entire grade, each worth 50 points. He doesn't drop any, but makes your lowest grade have half of the weight, so there's 625 total points.

2) Assignments. The wording on some of the problems were vague, to say the least. People kept on badgering him with questions about what he wants from us, and there were many games of mental tug-of-war over whether or not to include/not include some data in the problem, how we should present our solutions, etc. Luckily, collaboration was encouraged, so we could argue all we wanted in our GroupMe (trust me, we did).

3) He was pretty slow with releasing assignment grades, not too slow, just could've been faster.

I recommend taking this class, but getting a friend to collaborate with. Shouldn't be too hard if you have Python experience (I came in having taken CS 1110), he covers the basics for the first week, then goes into numpy and Pandas, and the last week was the god-awful dask package.

#tCFspring2021

Great class. Like others have said, you are kind of at the mercy of the TA grading your homework. I ended up going to office hours a lot so that my code was more or less exactly the solution they were looking for. It's also very helpful to connect with classmates so that you can collaborate on the homework. Don't wait until the last minute to get started on the homework. There's usually a few problems that require some serious time and thinking.

Overall, not a bad class. I don't particularly recommend it if you already have experience using Python because the first few weeks will be pretty boring, but the classes usually end early and your grade is made up of weekly homework assignments and a couple open-note, open-Internet exams which are not bad at all. My only issue with the class was the extremely arbitrary grading system. The homeworks were 15-20 questions but only 4 were actually graded and the points deducted were inconsistent and again, arbitrary, but this was probably more the TAs because Professor Holt would just recommend you appeal your grade. Professor Holt is awesome though and did a great job given the circumstances, so if you're willing to make sure you write the code exactly as they want it, it's a pretty easy class

This class is alright, but your grade doesn’t tend to reflect the work you put in. I’m not entirely sure the GPA counter is that accurate for this course as people I know are having a similar experience to mine. Holt is a nice guy but not the best teacher per se. Your grade is primarily composed of homework assignments but the grading is a bit off where TAs only grade about a 1/4th to 1/3rd of the questions and base your entire grade on those. It’s also weird because the questions being graded aren’t the same across the board (you get graded on hardest ones, friend gets graded on easiest ones). Tests in class weren’t necessarily hard but there wasn’t enough time to complete, let alone review. I’m not exaggerating when I say I went to Office Hours every week to get help/get answers check to ensure getting a good grade but still got points taken off even after having it looked at by TAs. OH are poorly managed in every aspect so don't expect a lot of help once the assignments get harder. I'll probably end up around a high B+ (normally I'd be thrilled) but I've put so much work in to just get that when it's really structural issues that inhibit my grade and not the material itself. If you need a stat elective, STAT 3080 is sooo much better structurally and if you want a good grade, at least that class will reward hard work you put into it. Not a hard class in terms of material, just annoying based off how it's structured. If you take this, become friends with the people in the class so you can at least have a buddy to check homework with.

Best class I have taken at UVA. It is so useful and Holt is the man! Can't recommend enough!

Professors Holt is great and I couldn't recommend this class more! For the Spring of 2019 he changed the course from the old 15 assignment format so that the grading consisted of 9 assignments and 3 exams. The first exam was an in class open internet, open notes exam that wasn't too challenging but the time constraint certainly made it difficult to complete the entire exam. Holt moved away from this format for the next two exams making them essentially the same as the homework assignments just longer and slightly more challenging. Another great thing about this course is attendance is far from necessary. Holt posts all his notes online and if you have taken a previous coding course (CS1100 perhaps) you will certainly not need to go to class everyday as we don't learn anything that challenging. All in all, I can't recommend this course enough I learned a lot and for any aspiring data analyst the applications of python that this class teaches are very useful.

Professor Holt is the nicest man you will ever meet in your life. The class however is kind of wonky. Spring 2019 he made the class 12 assignments and 3 exams which progressively got harder. At the beginning of the semester he said no loops, but then on the last exam and assignments all the questions required loops. However i don't think the class was that hard. Its a hard A but easy A-. you need a 95 to get an A. Overall good class and would recommend.

I highly recommend this class. It is an extremely interesting class if you are at all interested in statistics/computer science/data science and is extremely applicable in the real world. The course involves just 12 homework assignments, each of which covers slightly different topics (pandas, bumpy, statistics concepts, some data visualization). The homework start off easy but get pretty long and more difficult at the end (though never too bad). If you work with other students and go to the TA office hours, it is not hard to do well in this class. Finally, I will say this class was one of the most rewarding classes I have taken at UVA.

Great class, teaches you a lot about how to use python for data manipulation. The other comments here which are longer are accurate if you read those. If you are worried about it overlapping with intro CS it doesn't, the content differs and you will continue to learn new things. You learn new functions, new libraries (Pandas, Numpy, and Plotly) and also get experience coding using a different wrapper i.e. Spyder

I highly recommend this class to anyone who is interested. The class starts with basic coding know how and progresses into various data manipulation "puzzle" type assignments that pretty much boil down to manipulating data frames. The typical knock on a class like this is that everything could be learned via online tutorials and one would be better off taking a more rigorous class, but I still think that it is worth taking. The assignments are very well crafted and I now would feel very comfortable tackling a multitude data manipulation oriented problems.

To E-school kids: This is an introductory coding class that requires no prior experience. If you are taking this just to learn some Pandas/get an easy A then please don't complain about how easy the class is. No one cares.

This course is one of the best I've taken at UVA in terms of real-world applicability. Python is a super easy language to learn (coming from a guy with no coding experience before this class) and Professor Holt is a diamond in the rough in the Statistics Department. Course is 12 assignments, no exams and no final. You can work with others on the assignments but it is best to try to do the assignments on your own before getting help from others. It's basically like solving a puzzle, which I really enjoyed, and nothing is overly complex. One assignment towards the end of the course will be very tough, so be ready for that. Overall though, I decided to take this class on a whim and it ended up being my favorite class of the semester. Your grade will be exactly what you put into the course, but expect to work a fair amount to get that A. DON'T USE FOR LOOPS UNLESS HE SAYS YOU CAN USE FOR LOOPS. After like assignment 4 or 5, he bans for loops unless in very specific circumstances. Don't try to argue with him if you use one and he docks points. Get used to not using for loops, groupby and masking are honestly easier and faster. This is usually an issue with people who don't go to lecture, so go to lecture. Also, Professor Holt has people turn in assignments on Wednesdays in class. A significant number of people every week would come into the class halfway through Professor Holt's lecture, walk to the front of the class to turn in the assignment, then straight-up leave the class immediately. Not only is this fundamentally rude, it got on Professor Holt's (everyone else's) nerves. So, do yourself a favor and wait until the end of the class to turn it in if you absolutely need to turn in the assignment late for some reason. He's super chill about it and won't mark you off points. In fact, he'll respect you more.

This is a very heavily coding based class, so if you have any background in coding, it definitely makes it easier to learn and pick up things, but Holt does assume that some students are learning Python for the first time. This course focused more on data cleaning/manipulation so that it was be easier to use rather than statistical analyses. There was classwork due every week and homework every two weeks. Classwork was generally straightforward to the python examples he showed in class, and homeworks sometimes were more difficult, but graded pretty loosely.
Two midterm exams and one final group project. Exams were straightforward and similar to the practice exam that was posted on collab. The group project was the only groupwork during the semester and was the 'final'.

This class is definitely easy for someone who has experience with coding in any language, but might be tricky for someone with no programming background at all. The homework can be difficult some times but it is graded very leniently. I wish Professor Holt would make lecture some how required to attend since he is super nice and knowledgeable. Overall, not hard to get an A, and you will learn quite a bit about coding and statistics.

This is a new course that the Statistics Department offered this semester. I think they are offering this course since Python is language that is easy to learn and has useful statistical libraries that might help you in the real world. There were many bumps since it was the first semester Holt was teaching this new course, such as extensions and general expectations. I wish he can make the course better by making the homework questions much more clearer or by giving out the answers(output) first so that we know what we should expect in our coding, just like how it is done in CS 1110. I don't know when this will be offered again, but if you want to learn coding in Python and do statistical analysis, this class is for you.

If you have experience with Python from CS 1110, it is pretty simple for the first month or so. You will be better prepared than anyone else in the class. Taking CS 1110 before taking this class helped me immensely. I recommend taking CS 1110 first. After the first month, the class picks up the pace. You learn how to use statistical libraries such as Numpy, Pandas, DataFrames, and etc. You use these libraries to clean data, analyze data, and perform calculations by writing out code in Python. The last two months are quite challenging since you have to deal with massive data sets and the questions that Holt makes up is sometimes unclear. He is open ended in how you interpret the questions as the semester moves on, but as I said, it would be better if he made the questions more concise and clearer.

Overall, it is very similar to a CS course--homework is all coding and the exam is reading code and writing code. Pretty good class and you learn a lot but there are some homework that is insane and take a lot of time. Be ready to stare at the computer for hours which composes of looking at the given data sets, try different things by writing small bits of code and testing it.

Two midterm exams. The best way to prepare for exams are to look at his sample questions that are released before the exam. The exam is very similar to the sample questions. Make sure you know how to do all the sample questions, then you will at least get a B. There is a final group project. It really comes down how well your group is determined but I honestly am not a fan of group projects for some reason since I like to be in control of all the code and analysis. It's a personal thing, but I hope he gets rid of the final project with a final exam or move it to an earlier date where you don't have to be obsessed and stressed with the project and other final exams.

Overall, its not a bad course. I learned a lot. I had experience with Python in CS 1110 and did well in CS 1110, so it was easier to pick up the material and coding part. I may be a bit biased since I loved Python as my first programming language, but I loved learning how to use statistical libraries in Python by applying loops, conditionals that I learned in CS 1110 for doing statistical analysis. If you do not like coding, you might figure out that you should drop the course early on. It is very easy for the first month, but if you are struggling I recommend dropping it since you should be aware that the homework is not that simple as was in the first month since it involves a lot of effort and time to look at data sets and perform calculations by writing code.