Group 1: Code Review

[shrnkm]

Assignment 0: Peer review for sheet 01 [3 pts]

Did you do that? Cause I am not gonna review your reviews for my review on you code. This was not the way is supposed to be. So... good job guys, full points. 3/3 points

Assignment 1a: Importing a csv file [2 pts]

That would be nice if you use a smaller number of rows for your dataframe to be printed out, aesthetically speaking. Reviewers would need only a short look through the thing, like 15 lines or so. And I guess we should always check whether the path in which we wanted to save our data is available or should be created. Check the solutions. 2/2 points

Assignment 1b: Working with dataframes [2 pts]

Hmm! This NaN thing might actually be a better idea than removing 0 RTs. Did you talk about with tutors? Cause removing these 0 RTs made a big problem in some parts of the code. May I give you 3 out of 2 points, for that? Any tutor checking here... no? OK! I don't think it is a good idea to print all data in your code. For checking, yes, but don't forget to remove these check-print lines after you're done. 3/2 points

Assignment 2a: Boxplot and Violinplot [2 pts]

Who's the subject guy/gal 9004? Fast and accurate! Though your legend could be more informative. I cannot reference you to solution in this case, cause, I think, those Go/No-Go labels are not informative (in fact even wrong) as well. The reason: they did not use your NaN idea! They just removed all RT=0. 2/2 points

Assignment 2b: Violinplot combining all data of all groups [3 pts]

So you know you did this part wrong already, right? There's a trick for no-value x-axis you can find on solutions, and your four plots are working with RT (no accuracy check). But you did try. So we will take that extra point back. 2/3 points

Assignment 2c: Scatterplot [1 pts]

In fact the plot shows, that a high accuracy agrees with a higher RT,

No, it does not! [please imagine a surprised-denial emoji, here] Your regression line is a flat one, guys. No correlation then! The whole story of regression is being told by the slope of that vicious line. Anyway, since I feel something is wrong with this experiment analysis design, like that RT=0 removals, and cause you used that regplot thing and stuff, we are cool.

1/1 points

Final Grade: 13/13

[DorianDaimer]

Two small remarks concerning general coding style:

• you rely heavily on for loops for reading the data files. This could have been solved more cleanly with pandas and glob/os functions. Check the solution for a neater way of doing this.
• due to the use of for loops you have to initialize some lists to append to them. This could be avoided if you would use less for loops Overall this does not impact the functionality of the code and otherwise it was nicely commented and clearly structured. 2/2 points