TA comments

[ChadFibke]

Coding Style

The good :thumbsup:

• Most target names and objects created in R script are informatively named! This helps to inform the reader what is being made!

• Code looks great!

Needs work :thumbsdown:

• Add some comments to your Make file! This will help annotate what is going on!

• Some objects are poorly named... 1f, 2f, 3f are not informative!

Coding Strategy

The good :thumbsup:

• Wow factor: Extremely High level analysis preformed, integrating almost all class material learned :star: :star:

• make clean removes all the output! This allows us to see how your pipeline can handle the input and scripts!

• Running make all configures all output and finishes without any major error!

• The report was curated with r markdown syntax for additional structure! :star:

• Pipeline created meaningful intermediate files that can be used later on! For example, what If I wanted to use the common_words.tsv for evil? With your pipeline I can.

Needs work :thumbsdown:

• Instead of adding these extra instructions:

render_cloud_plot.pdf: files/cloud.pdf
    cp $< $@

render_bar_plot.png: files/bar_plot.png
    cp $< $@

could you not just use the relative pathways to cloud.pdf and bar_plot.png when using them as dependencies for your final report? These steps seem to be unnecessary, and create more unsorted files in your repo!

• Try making a log file. This text file can detail the process of a pipeline, add information from each step, and can allow the user to read the process of a large pipeline! This can be done by:

rule 1
 file("test.log")

rule 2
cat("\n Your result for step 1", file = "test.log", append = TRUE)

rule 3
cat("\n Your result for step 2", file = "test.log", append = TRUE)

• In order to show that the Makefile is responsive, not hard coded for the exact input, it would be a good idea to add 2 data dependencies like half_of_donalds_words.text and all_donalds_words.text. This could show that the pipeline creates the desired output with 2 different starting points! Also this could have been put into separate analysis that can be run separately using different phonies!

Creativity & Ease of Access

The good :thumbsup:

• Easy navigation from canvas to your repo!

• Thanks for provided a detailed README describing what was changed, commenting on the repo contents, and giving detailed instructions on how to install the dependencies!

• Wow factors: use of multiple languages, adding a pipeline figure for others to view, and added multiple new R scripts!

Needs work :thumbsdown:

• No need to save your .Rhitory file. In the future add this file name to you your .gitignor file.

• Use the mkdir command to help organize your repo. As of now many files are out in the open, and has a disorganized look. For example you could use mkdir data to store all your csv files, and make R to store all your r scripts! This directory/folder setup could be part of the instructions for one of the initial/last rules in your pipeline.

# For mac

mkdir data R whatever_you_want

• When running the pipeline with your instructions resulted in some missing dependencies like tidytext and repurrrsive. I see you get the user to install wordcloud and other packages, but make sure to provide an exhaustive list!

• Also I could only run the python script when I used an absolute pathway, not a relative pathway. This could have been a "me issue", so I wont doc you marks.

[rasiimwe]

Hi @ChadFibke

Thank you very much for reviewing my work and for this comprehensive review and helpful comments - it's like another tutorial all together 😄 👍 👏 👏 Thanks . I will apply these in my subsequent R projects.

Many thanks! Rebecca

[ChadFibke]

Hey @rasiimwe,

I would also like to thank you for your effort you put into every assignment, and every peer review! :star:

Cheers, Chad

[rasiimwe]

Thanks @ChadFibke 😄