The functions contained in this folder surround the implementation of various cross validation procedures, both time series and non time series in nature. The goal of this library is to make such procedures available to kdb+ users using a q-like syntax.
Within this folder, users will find xval.q which contains base-level implementations of cross validation procedures, grid/random/Sobol-random hyperparameter searching methods and multi-processing procedures.
- embedPy
- sobol-seq
The Python dependencies for the FRESH library can be installed by following the instructions laid out in the ML-Toolkit level of this library. Note that sobol-seq must be installed using pip.
Place the ml library in $QHOME and load into a q instance using ml/ml.q
The following will load cross validation functionality into the .ml namespace
q)\l ml/ml.q
q).ml.loadfile`:xval/init.qDocumentation is available on the Cross Validation homepage.
The cross validation library is still in development. Further functionality and improvements will be made to the library on an ongoing basis.
If you have any issues, questions or suggestions, please write to ai@kx.com.