build_full_trees_as_named_tree_mats_2019_05_27.m

options.whole_brain_h5_p_map_file_path = '/nrs/mouselight-v/cluster/classifierOutputs/2019-09-06/whole-brain-p-map-as-h5/2019-05-27-whole-brain-p-map.h5' ;
options.whole_brain_h5_p_map_dataset_path = '/prob0' ;
options.whole_brain_h5_p_map_properties_group_path = '/prob0_props' ;
options.skelfolder = '/groups/mousebrainmicro/mousebrainmicro/cluster/Reconstructions/2019-05-27/skeletonization' ;

options.outfolder = '/groups/mousebrainmicro/mousebrainmicro/cluster/Reconstructions/2019-05-27/build-brain-output/full-as-named-tree-mats' ;

options.writefull = 1 ;
options.writefrag = 1 ;

options.viz = 0 ;
options.debug = 0 ;

options.sizethreshold = 100 ;
% decide crrop size from data as function of chunksize%cropSize = 1000

% overlap of running windows on skeletonization
options.fullh = 15 ;

%    lengthThr = 10;% (um) lengthThr for each leaf branch to decide to prune
%    largesampling = 200;%(um) make it more sparse for larger segments
options.lengthThr = 10 ;
options.largesampling = 200 ;
% sampling options are 'uni' [default] for uniform sampling (for 3D) and 'curv' for curvature weighted sampling (for 2D)
options.sampling = 'uni' ;
options.samplingInterval = 5 ;

%post segmentation parameters
options.tag = '' ;
options.prune = 1 ;
options.filterGraph = 0 ;
options.graph2branch = 1 ;

options.maximum_core_count_desired = 20 ;

build_full_trees_as_mats_workflow1(options) ;