search

Tpatni719 / gsMAMS

GNU General Public License v3.0
0 stars 1 forks source link

JOSS Review: Functionality: Have the functional claims of the software been confirmed? #6

Closed njtierney closed 5 months ago

njtierney commented 6 months ago

I think that to demonstrate the functional claims of the software there should be a short form vignette which describes how this software solves a problem, and work through how you would expect a user to use this software.

See the vignettes chapter in the R packages book for more details on this.

Review link: https://github.com/openjournals/joss-reviews/issues/6322

Tpatni719 commented 6 months ago

In the paper, we have provided comprehensive examples explaining each argument of every function so that the user can easily use the package based on these examples.

njtierney commented 6 months ago

I think that the paper should be turned into a vignette as well. The paper.md will not be shipped with the package.

In the paper, we have provided comprehensive examples explaining each argument of every function so that the user can easily use the package based on these examples.

Have you read the vignette chapter of the R Packages book? This provides examples of how to create a vignette. In addition this will provide the output of the code that you use.

Tpatni123 commented 6 months ago

Thanks for the link! I will add the vignette.

Tpatni719 commented 5 months ago

I have added the vignette.

Tpatni123 commented 5 months ago

And if in case you might want to create the vignette, it will take a LOT of time because the number of simulations is very high. So, I just wanted to give you a heads-up.

njtierney commented 5 months ago

Hi there,

For the vignette, you could make this run faster by changing the number of simulations to be lower, and then reference this in the code that users should run a certain number of simulations.

njtierney commented 5 months ago

For reference it took about 5 minutes on my M1 laptop to render the vignette.

Tpatni123 commented 5 months ago

Thanks for letting me know! I think 5 mins is fine and I am hesitant to reduce the number of simulations as it will change all the results and as a result, the type I and type II errors will not reach the desired level. So, I think the current number of simulations is fine.