--set_min_feat_sum

Switch

--set_min_feat_sum N

Description

The minimum total occurrence a feature must have, summed across all groups, to be kept — an alternative/companion to --set_p_occ's percentage-of-groups threshold.

Argument and Default Value

An integer. Default: 0 (dlac.DEF_MIN_FEAT_SUM) — no filtering.

Details

Marked #TODO: Deprecate? in the source next to its add_argument call, for what that's worth. Used together with --feat_occ_filter and --set_p_occ — passed as fr.createTableWithRemovedFeats(pocc, minfeatsum, groupfreqthresh, ...), i.e. both thresholds are applied together when refining a feature table.

Other Switches

Required Switches:

• --feat_occ_filter

Optional Switches:

• --set_p_occ

• --group_freq_thresh

Example Commands

dlatkInterface.py -d dla_tutorial -t msgs -c user_id -f 'feat$1gram$msgs$user_id$16to16' \
--feat_occ_filter --set_min_feat_sum 50