Hopefully will remove the data overwrite issue

May fix the possible overwrite error

Correct bugs when attempting to allow additional features when max_cor < 1

use stand_val everywhere
correct condition for pearson

Fails when tested thought

Makes some of the comparisions easier

Postfix was done correctly, issue was sorting of the phi_temp from the file

More features will be allowed this way

Since MPI is not being used in the runners

Testing the phi ouput scheme

Subtract and less then 1e-10

Rinitialization removed existing data leading to the problem

Test in the docker now

fix that compilation error

Also bug fixes associated with those changes

Complining should work now

Now to clean up

Now need to adapt the python interface to match

Classification scores now positive in feature space output file

Similar concepts get the same definition

Comment out blank lines

wrong comparison in an is_valid function

This seems to not be feasable with OpenMP

Everything seems to pass now

SIS also works with multi-threading

It seems to work

for paramLogNode _sign_alpha was never initialized

If they are then remove them from feature set

check if 4.5 directives can be used

This was throwing seg faults if it was not monotonic

_reparam_phi now seems to work correctly

Index bugs and array filling issues fixed

Actually create new features with the reparameterized nodes