StatisticsHealthEconomics / multimcm

Bayesian relative mixture cure modelling with Stan.
https://StatisticsHealthEconomics.github.io/multimcm
Other
0 stars 0 forks source link

Large files? #2

Open giabaio opened 3 years ago

giabaio commented 3 years ago

Hi @n8thangreen --- I was trying to catch up and pull the latest changes you made, but it failed because there is a bunch of very large files. I think I had this problem with survHE too, because when you build rstan locally, it compiles the C++ code and creates a bunch of .o files --- specifically in this case:

src/RcppExports.o
src/stanExports_exp_relative_mix.o
src/stanExports_gompertz_relative_mix.o
src/stanExports_weibull_relative_mix.o

My solution was to add these to the .gitignore file:

# Large files generated by Stan
src/*.so
src/*.o
src/stan_files/*.hpp
src/stan_files/*.o

(though I think the folder where these get stored for you is different...). Does this make sense?

n8thangreen commented 3 years ago

Hi, thanks for looking at this. I have included these in .gitignore already but I think the problem is the history is still there with these files. The output from stan is also pretty big in the data/ folder. I can wipe files from the history like this: https://docs.github.com/en/free-pro-team@latest/github/authenticating-to-github/removing-sensitive-data-from-a-repository

So if you think I need to do this then I'll have a go...

giabaio commented 3 years ago

Isn't a simple(r)

git rm --cached path_to_files/*.o

do? (I think what that does is remove the files on the remote)

As a general thing, it's not a massive problem --- but I think it doesn't let me pull your changes to my local repo, because the files to transfer are too big?