package linwrap
Install
Dune Dependency
Authors
Maintainers
Sources
sha256=398405fa76a86cd7389bae970d892c506da71adcd4b53ef89f4d6003da28049f
md5=c7c8919f14df973e512a8d1de21a8924
Description
Only L2-regularized logistic regression is supported currently. When using bagging, each model is trained on balanced bootstraps from the training set (one bootstrap for the positive class, one for the negative class). The size of the bootstrap is the size of the smallest (under-represented) class.
usage: linwrap -i : training set or DB to screen [-o ]: predictions output file [-np ]: ncores [-c ]: fix C [-w ]: fix w1 [-k ]: number of bags for bagging (default=off) [-n ]: folds of cross validation [--seed ]: fix random seed [-p ]: training set portion (in [0.0:1.0]) [--train <train.liblin>]: training set (overrides -p) [--valid <valid.liblin>]: validation set (overrides -p) [--test <test.liblin>]: test set (overrides -p) [{-l|--load} ]: prod. mode; use trained models [{-s|--save} ]: train. mode; save trained models [-f]: force overwriting existing model file [--scan-c]: scan for best C [--scan-w]: scan weight to counter class imbalance [--scan-k]: scan number of bags (advice: optim. k rather than w)
Published: 22 Jul 2020
README
linwrap
wrapper on top of liblinear-tools
Dependencies (11)
-
parany
>= "11.0.0"
-
minicli
>= "5.0.0"
-
dune
>= "1.10"
-
dolog
>= "4.0.0" & < "5.0.0"
- dokeysto_camltc
-
cpm
>= "5.0.0"
- conf-rdkit
- conf-python-3
- conf-liblinear-tools
- batteries
- base-unix
Dev Dependencies
None
Used by
None
Conflicts
None