factorMerger

factorMerger copied to clipboard

> devtools::install_github("ModelOriented/factorMerger", build_vignettes = FALSE) Downloading GitHub repo ModelOriented/factorMerger@HEAD Error: Failed to install 'factorMerger' from GitHub: input string 110 is invalid In addition: Warning message: In gsub("[^/]", "", file_list) :...

dar143

Meaning of numbers tree

1

Hi, I've read the [documentation](https://arxiv.org/pdf/1709.04412.pdf) and I'm not able to understand the meaning of numbers appearing in each tree leaf... In the example within the documentation (figure 2) appears that:...

jgarces02

I wonder what's he doing when computations take more than 10 minutes. Whether he's still on the first iteration or it's in the middle? Are there any time estimates on...

MarcinKosinski

at least for `plot.factorMerger`, `print.factorMerger`, `cutTree`, `pisa2012`

pbiecek

?factorMerger is not working right now

pbiecek

is the whole knitr needed for kable() function?

1

without the `kable` used here https://github.com/MI2DataLab/factorMerger/blob/c8c3c486b994dcad0499f0de12f4ad7cc4c9d944/R/factorMergerClasses.R#L212-L228 `knitr` could be moved from `Imports` to `Suggests`, hence final user doesn't have to rely on the `knitr` version to use the `factorMerger`. What...

MarcinKosinski

This 1-liner https://github.com/MI2DataLab/factorMerger/blob/master/R/factorMergerClasses.R#L6 drives the final user to install an extra package. There is no point to import pipe operator from the extra package when you have one in the...

MarcinKosinski

reshape2 namespace has many conflicts with the dplyr package so I advise not to encourage users to use reshape2 anymore, so don't make them install this package because you just...

MarcinKosinski

just import full dplyr and ggplot2

1

Documentation isn't cleaner and easier to read, nor NAMESPACE where you import almost every single possible function from `dplyr` or `ggplot2` https://github.com/MI2DataLab/factorMerger/blob/master/NAMESPACE#L26-L82. I recommend just to simply roxygenize the ```{R}...

MarcinKosinski

Cluster coloring with non-equidistant nodes

2

There is a bug in the coloring (`plot.factorMerger` with the argument `statistic = "p-value"`). See below:  Needs to be fixed.

durszlaczek