Both convolution and cross validation have been parallelised in parallel_convolutional_sparseFiltering.py. This uses all available cores. to control the number of cores used change the -s argument for the convolution step. E.g. if there are 20 cores and you wish to use 10 then the -s argument should be then number of features to convolve divided by 10, in which case -s should be 40.
There is currently no easy way to control the number of cpus used for cross validation. It will use the length of the grid to search multiplied by the number of folds number of cores.
Both convolution and cross validation have been parallelised in parallel_convolutional_sparseFiltering.py. This uses all available cores. to control the number of cores used change the -s argument for the convolution step. E.g. if there are 20 cores and you wish to use 10 then the -s argument should be then number of features to convolve divided by 10, in which case -s should be 40.
There is currently no easy way to control the number of cpus used for cross validation. It will use the length of the grid to search multiplied by the number of folds number of cores.