CSE 160 - HW7 - Advice from 15sp students: "What do you wish you had known before you started? What would you do differently? What advice would you offer to future students?" -------------------------------------------------------- Part I Advice: Be creative Iterate through this assignment. Each time through make it more detailed until you have an outline that you can start coding from like the previous homework assignments. Don't be afraid to revise and complete motivation and the structure of project. We have been tweaking our research questions as we go now that we have a better understanding of our data set, so it would have been useful to have down that work in part 0. start this project early. This is really useful to do before starting the assignment, since we hadn't come up with the minor details of the assignment yet, just the big picture. if you did part 0 thoroughly then it shouldn't take long This time around, I don't really regret anything. If I had to choose what I wish I had known before I started though, I suppose I wish I knew that I had a long road ahead of me. This would have motivated me to finish Part 1 a bit sooner so I could get started immediately on Part 2. I would not have turned in the assignment 1 day late for falling asleep all day yesterday, even though I have late days. I would advise future students to really take the opportunity to do the project on something that is interesting to them, no matter how obscure the topic may seem to others. I wish I had made better use of office hours to plan with people who could give me advice before I launched headlong into an ill-concieved idea. Initially in part 0 of this assignment, when I found the data, I thought I would be able to do some sort of data analysis on it, but as I thought more about it, the data I had didn't seem to be a good set to produce meaningful results. What I would do differently is look for a broader range of data sets out there. I would suggest to start early when it comes to looking for the set of data that would best fit your interests as well as what you are looking to do for the analysis. Partners need to make sure they understand each other's function and role. Do not misread the function and purpose of task. I think this assignment is straight forward. I wish I had know what dataset I wanted to use. The largest bulk of the time I've spent so far on this assignment was just trying to figure out what to do and where to get a dataset. More guidance towards datasets might've been nice. I would suggest future students to figure out what they want to do while they search for data sets. nothing really. Start early Get all the data first or you will change your questions frequently. Finding good data sources can be difficult. Search around on other websites as well as the ones provided in the assignment I think the best way to start, is start the scratch part of the program, and get he program to begin reading the file you will be using. And start organizing the code. I made sure I had my ideas down a few nights before the deadline, because you don't want to cram for the project, if you want to do well. Learn from your previous assignment, and try to think about your mistakes for past projects, before you begin the final project. The workplan is the most important part. I also think that there needs to be a "questions" section, where students can put down the questions they may need to have a TA answer. Not that the TA will answer them when the assignment is turned in, but to identify a student's own shortcomings and what they need to work on. -------------------------------------------------------- Part II Advice: Under the data, how it is structured and on paper figure out the best way to get from the data inputs to the output. Some coding decisions I made were because that was the only way I know how to do it but I know on some things there has got to be a better way. AND that is why you start EARLY! So you can go to TA office hours and ask these questions!! Never let your project too difficult and made your research questions as directly as possible. how much time it would take I wish I had known that the simplicity with which I could describe the tasks of each portion of the project does not make it reasonable to underestimate the difficulty of actually coding them. I would have probably tried to find a partner. I would mainly advise students to do the project on something which interests them, so at the very least, if the project kicks your butt, you can enjoy it more than if the actual subject matter is boring to you. chose right scientific question is more important to get thing right. plan on smaller, simpler data set I should've know about how coefficient correlation work beforehand. How to properly go about answering my question. Initially, the approach might seem to be straightforward, but depending on the data file or what you can do with Python you might come across some unexpected problems. I would suggest starting early, though you might have developed the question and have a plan on how you would approach this task, there might be some issues you might encounter that will prolong the time it takes for you to finish. I wish I had known how to clean data more efficiently. Also, I would document all of our work and give someone the steps to see if the outcome is easily reproduced. Work on your project ahead of time. Try to get your code finished and polished two days before the deadline, so that you have time to go over it and write your report. Note, you won't be able to finish your report, unless you finish your project. I would have tried to read my files before I actually started coding my project, to see what file errors I would run into. I wish I had known about MacPorts and clients to aid in building/installing library packages The things that became problems for us were not foreseeable before we started. I wish that I had realized how strict our filters were going to be. I would've started with less filters. Also, I would have put more time into the user interface. I would advise students to think about design much more than we did. I wish we had defined our functions and what we were trying to do a little better, since, as mentioned above, we ran into problems where we went sure we were actually doing what we intended to do. Advice: Start looking for data sets early. Run some small programs on them to make sure your end goal is feasible with that data set. START EARLY! This definitely took me a lot of time to complete. I wish I had known more about importing modules. -------------------------------------------------------- Part III Advice: The importance of being able to succinctly describe a project I learned how to carefully read package documentation and installation to fulfill specific requests of our interest I wish I did understand the coding concept a little better. My partner understood some of the concepts a little better and he was of great help when I would ask about certain parts of the program and when it came to brainstorming what we wanted to do. Do a better estimation about the time we need. Don't stress too much! Advice to other students: relax, take a deep breath and drink a mouthful of water before your present. My data had over 200 points of result so I had difficulties choosing what was significant to present. I think this would have been prevented if I had carefully planned and detailed what I thought was the important parts. Everything went smoothly for creating the slides! I would suggest having slides that convey a good amount of info since you're limited to only 5. Again, with the project in general, I suppose I just wish I had known how difficult the actual coding and function implementation would be. I would have studied more and made sure I was confident I had the knowledge necessary to make the each part of the program and the knowledge of what each step entailed in terms of code. I would advise students to know their current state of comfort or skill with coding and choose an interesting project within those limits or slightly beyond them. Otherwise, the task can be extremely overwhelming. be able to condense alot of information to few slides how to find and learn about different Python packages you can find online. It's difficult to explain scientific concepts quickly to people who are unfamiliar, so it was hard to summarize topics. As I worked with images, it was really hard to debug and find out my mistakes. I had to look at so many numbers and find out the difference. I wish I had known to start from a smaller dataset, even if I am using images as my data. I should have chosen smaller picture with less pixel values at the begining. Two minutes is not a lot of time. Focus on what is the most important part of your code ahead of time, and make sure you can explain it very efficiently. This likely means writing down your logic as a plain English explanation as you go if you don't already do this. This part of the project was pretty straight forward, especially after the writeup in partII. I wish I had known how short the presentation was sooner. --------------------------------------------------------