0todd0000 / mwarp1d Goto Github PK

Dear Todd,
when I try to execute the export of the warped data into a matlab .mat file the software crashes and I get the following error command:

Traceback (most recent call last):
File "C:\ProgramData\Anaconda3\lib\site-packages\mwarp1d\ui\widgets\menubars.py", line 44, in on_export_mat
self.panel.data.write_mat(fname)
File "C:\ProgramData\Anaconda3\lib\site-packages\mwarp1d\ui\dataio\data.py", line 104, in write_mat
savemat(fname, self.get_dictionary(format='matlab'))
TypeError: get_dictionary() got an unexpected keyword argument 'format'

I use windows 10 with matlab 2019a if this will in any way help you to figure out what causes the crash!

thank you for your efforts
Sincerely
Andy

Hi,

On the website, it is stated that you can use ttests2 with Bonferroni correction for ANOVA post hoc analysis.
My question is, should I use ttests2 or ttest_paired for repeated measures ANOVA?

Thanks

The minimal JOSS requirements testing reads:

Documented manual steps that can be followed to objectively check the expected functionality of the software (e.g., a sample input file to assert behavior)

I guess you could basically argued that some of the steps in the documentation full fills this requirement. But there is not really any objective criteria to say if the warping was correct. In other words how can I know the library does the same as the last release.

I think you could benefit from automated testing. Tests are also a nice if you ever get contributions from others as it helps verify their code. Using something like "Github actions" is very easy to setup,

This not really relevant for the JOSS review, but just a hint.

If you create the conda package with the noarch option, the package will work on all platforms by default. Hence, you easily get linux support, without creating a conda-forge package.

Where is the source code for the documentation? I think it would be good practice if the sphinx source code for the documentation lived in the repository.

Contributing to documentation is also big part of "open source", and it would allow you automatically build the documentation with a CI service like github actions.

The application uses dependencies which are not listed setup.py or in the conda recipe.

It think all the formal JOSS requirements for documenting contributing guidelines, license etc. are all fulfilled in the documentation page.

However, I would still suggest you make some of these things a little easier to find by extendíng the readme page in the main repository. Many users will probably find the software first on GitHub, and would be nice if the readme was more informative.

See these examples for awesome readme files:
https://github.com/matiassingers/awesome-readme