Parameters

Note

Siren allows to tune various parameters via an interface menu Edit ‣ Preferences.

Non-default parameters can be stored in a preferences file, e.g. via Export ‣ Export preferences.

Parameters are listed below by their label and identifier, as used respectively in the interface and in preference files, with a short explanation of the parameter, type, default value and indication of file where the parameter is defined.

The parameters that can be set through the preference interface can be mainly divided into two categories, mining parameters, that pertain to the mining algorithm, and visualization parameters, that pertain to the interactive visualizations, each split over several tabs.

Mining parameters

Mining parameters are specified here and can be set through the interface.

Quite many parameters are available for tuning, but in general only a few of them need to be adjusted. In particular, the parameters setting threshold values on the support of the queries, on the accuracy of the returned redescriptions, as well as on the type of queries, are the most important ones.

Mining: Method

Mining algorithm

mining_algo

• Algorithm to use for mining, greedy ReReMi or based on classification trees (restrictions apply).
• single option in {reremi, layeredtrees, splittrees, cartwheel, fim}
• default: reremi
• miner_confdef.xml

Mining: Queries

LHS max number of variables

max_var_s0

• Maximum number of variables in a left-hand side query, maximum tree depth in tree-based algorithms.
• open text
• default: 4
• miner_confdef.xml

RHS max number of variables

max_var_s1

• Maximum number of variables in a right-hand side query, maximum tree depth in tree-based algorithms.
• open text
• default: 4
• miner_confdef.xml

Mining: Candidate filtering

Min contribution

min_itm_c

• Minimum contribution, number of supporting entities impacted by the addition of a literal in ReReMi or minimum node size in tree-based algorithms. Values between 0 and 1 are interpreted as ratio of the total number of entities.
• open text
• default: 3.0
• miner_confdef.xml

Min support

min_itm_in

• Minimum number of supporting entities. Values between 0 and 1 are interpreted as ratio of the total number of entities.
• open text
• default: 0.05
• miner_confdef.xml

Min uncovered

min_itm_out

• Minimum number of uncovered entities. Values between 0 and 1 are interpreted as ratio of the total number of entities.
• open text
• default: 0.05
• miner_confdef.xml

Min improvement

min_impr

• Minimum improvement score for an acceptable extension.
• open text
• default: 0.0
• miner_confdef.xml

Max initial candidates

max_inits

• Maximum number of initial candidates explored, pairs for ReReMi, initial targets for trees. -1 leads to expand all.
• open text
• default: 100
• miner_confdef.xml

Initial candidates productivity

inits_productivity

• Productivity when creating an initial pair of literals from non-Boolean variables.
• single option in {low, medium, high}
• default: medium
• miner_confdef.xml

Excluded initial cand. pair types

inits_types_exclude

• Excluded variable types in initial pairs. B, C and N stand for Boolean, Categorical and Numerical variables, respectively. Note order is simpler type first, not by sides.
• multiple options in {BB, BC, BN, CC, CN, NN}
• default: {}
• miner_confdef.xml

Mining: Final filtering

Min support

min_fin_in

• Minimum number of supporting entities, applied for final filtering. Values between 0 and 1 are interpreted as ratio of the total number of entities.
• open text
• default: -1.0
• miner_confdef.xml

Min uncovered

min_fin_out

• Minimum number of uncovered entities, applied for final filtering. Values between 0 and 1 are interpreted as ratio of the total number of entities.
• open text
• default: -1.0
• miner_confdef.xml

Min accuracy

min_fin_acc

• Minimum accuracy, applied for final filtering.
• open text
• default: 0.01
• miner_confdef.xml

Max p-value

max_fin_pval

• Maximum p-value, applied for final filtering.
• open text
• default: 0.01
• miner_confdef.xml

Mining: Additional filtering

Max Area Overlap

max_overlap_area

• Two redescriptions whose area overlap is beyond this ratio are considered redundant.
• open text
• default: 0.5
• miner_confdef.xml

Max Rows Overlap

max_overlap_rows

• Two redescriptions whose row overlap is beyond this ratio are considered redundant.
• open text
• default: 0.5
• miner_confdef.xml

Mining adv.: Process

Nb processes

nb_processes

• Maximum number of mining processes to use in parallel.
• open text
• default: 1
• miner_confdef.xml

Log verbosity

verbosity

• Verbosity of the mining process output.
• range in [0, 10]
• default: 4
• miner_confdef.xml

Debug checks

debug_checks

• For debugging, double check counts.
• yes/no
• default: no
• miner_confdef.xml

Mining adv.: Other common parameters

Batch output

batch_out

• Maximum number of redescriptions returned from an expansion batch.
• open text
• default: 1
• miner_confdef.xml

Min total number of variables

min_fin_var

• Maximum number of variables per redescription, applied for final filtering.
• open text
• default: 2
• miner_confdef.xml

Weight Inside

in_weight

• Weight for entities inside selected area (only applies when such an area is specified).
• open text
• default: 10
• miner_confdef.xml

Weight Outside

out_weight

• Weight for entities outside selected area (only applies when such an area is specified).
• open text
• default: 1
• miner_confdef.xml

Var group compatibility

var_compat

• Fixes whether group ids should differ or match for variables to be considered compatible (gid=-1 is universally compatible).
• single option in {differ, match}
• default: differ
• miner_confdef.xml

p-value formula

method_pval

• Formula used for computing the p-value.
• single option in {marg, supp, over}
• default: marg
• miner_confdef.xml

Missing values belief

parts_type

• Type of Jaccard index, determines how to consider missing values when computing support statistics.
• single option in {rejective, optimistic, pessimistic, positive, negative, basic, exclu}
• default: rejective
• miner_confdef.xml

Mining adv.: Tree-based algorithms

Max rounds in CartWheel

max_rounds

• Maximum number of alternations in CartWheel algorithm.
• open text
• default: 10
• miner_confdef.xml

Split criterion

split_criterion

• Score used to evaluate splits during decision tree induction.
• single option in {gini, entropy}
• default: gini
• miner_confdef.xml

Mining adv.: Splittrees variant

Init depth

splittrees_init_depth

• Maximum tree depth for the split in the first round.
• open text
• default: 1
• miner_confdef.xml

Update target

splittrees_update_target

• Update the target between tree growing rounds.
• yes/no
• default: no
• miner_confdef.xml

Both sides change

splittrees_both_sides_change

• Require both sides to be different in successive tree growing rounds.
• yes/no
• default: no
• miner_confdef.xml

Dynamic side

splittrees_dyn_side

• Choose the side on which to grow the tree first depending on the source of the target (else always start with RHS).
• yes/no
• default: yes
• miner_confdef.xml

ReReMi: Queries

Single side OR

single_side_or

• Restrict use of disjunction to only one side.
• yes/no
• default: yes
• miner_confdef.xml

Multi categories

multi_cats

• Allow combining multiple categories of a categorical variable.
• yes/no
• default: no
• miner_confdef.xml

Conditional

add_condition

• Allow making conditional redescriptions, if suitable data is provided.
• yes/no
• default: no
• miner_confdef.xml

LHS Literals Bool

neg_query_s0_1

• Types of literals allowed in the left-hand side queries for Boolean literals.
• multiple options in {positive, negative}
• default: {positive}
• miner_confdef.xml

LHS Literals Cat

neg_query_s0_2

• Types of literals allowed in the left-hand side queries for categorical literals.
• multiple options in {positive, negative}
• default: {positive}
• miner_confdef.xml

LHS Literals Num

neg_query_s0_3

• Types of literals allowed in the left-hand side queries for numerical literals.
• multiple options in {positive, negative}
• default: {positive}
• miner_confdef.xml

LHS Operators

ops_query_s0

• Type of operators allowed in the left-hand side queries.
• multiple options in {conjunction, disjunction}
• default: {conjunction, disjunction}
• miner_confdef.xml

RHS Literals Bool

neg_query_s1_1

• Types of literals allowed in the right-hand side queries for Boolean literals.
• multiple options in {positive, negative}
• default: {positive}
• miner_confdef.xml

RHS Literals Cat

neg_query_s1_2

• Types of literals allowed in the right-hand side queries for categorical literals.
• multiple options in {positive, negative}
• default: {positive}
• miner_confdef.xml

RHS Literals Num

neg_query_s1_3

• Types of literals allowed in the right-hand side queries for numerical literals.
• multiple options in {positive, negative}
• default: {positive}
• miner_confdef.xml

RHS Operators

ops_query_s1

• Type of operators allowed in the right-hand side queries.
• multiple options in {conjunction, disjunction}
• default: {conjunction, disjunction}
• miner_confdef.xml

ReReMi: Initial pairs

Init min score

init_minscore

• Minimum score for an initial candidate to be considered.
• open text
• default: 0.01
• miner_confdef.xml

Pairs-expand balance

pe_balance

• Balance between generating initial pairs and expanding them (0 means generating all pairs first).
• range in [0, 10]
• default: 4
• miner_confdef.xml

Expansion order

pair_sel

• Order in which the initial pairs are processed, either overall next best score or alternating between side considering next best variables in turn.
• single option in {overall, alternate}
• default: alternate
• miner_confdef.xml

ReReMi exp.: Candidate expansion

Accuracy improvement coefficient

score.impacc

• Coefficient of the accuracy improvement for scoring extension candidates.
• open text
• default: 1.0
• miner_confdef.xml

Relative Accurracy improvement coefficient

score.rel_impacc

• Coefficient of the relative accuracy improvement for scoring extension candidates.
• open text
• default: 0.0
• miner_confdef.xml

Redescription p-value coefficient

score.pval_red

• Coefficient of the redescription p-value for scoring extension candidates. Negative values are interpreted as a threshold weighted by p-value factor.
• open text
• default: 0.01
• miner_confdef.xml

Query p-value coefficient

score.pval_query

• Coefficient of the query p-value for scoring extension candidates. Negative values are interpreted as a threshold weighted by p-value factor.
• open text
• default: 0.01
• miner_confdef.xml

p-value factor

score.pval_fact

• Weight for p-value in case of using thresholds for scoring extension candidates.
• open text
• default: 1.0
• miner_confdef.xml

Batch capacity

batch_cap

• Maximum number of candidates expanded at each step of the search.
• open text
• default: 4
• miner_confdef.xml

Amnesic

amnesic

• Not amnesic means the algorithm remembers previously generated redescriptions and avoid generating them again.
• yes/no
• default: no
• miner_confdef.xml

ReReMi exp.: Numerical attributes discretization

Max number segments

max_seg

• Maximum number of segments, i.e. distinct values for a numerical variable.
• open text
• default: 20
• miner_confdef.xml

Lower tail aggregate

lower_tail_agg

• Threshold for the lower tail, when aggregating keeping only extreme values.
• open text
• default: 0.0
• miner_confdef.xml

Upper tail aggregate

upper_tail_agg

• Threshold for the upper tail, when aggregating keeping only extreme values.
• open text
• default: 0.0
• miner_confdef.xml

Max aggregation

max_agg

• Maximum number of buckets aggregated together.
• open text
• default: 15
• miner_confdef.xml

Max product buckets

max_prodbuckets

• Maximum product of the number of buckets on either side.
• open text
• default: 5000
• miner_confdef.xml

Visualization parameters

Visualization parameters are specified here and can be set through the interface. For instance, they allows to modify the appearance of the maps, choose the colors of the different support subsets, etc.

Views: Vizualizations

Intab grid rows

intab_nbr

• Number of rows of the grid for visualizations in the tab.
• open text
• default: 1
• views_confdef.xml

Intab grid cols

intab_nbc

• Number of columns of the grid for visualizations in the tab.
• open text
• default: 2
• views_confdef.xml

Dot size

dot_size

• Size of the dots in points.
• range in [0.0, 10.0]
• default: 3.0
• views_confdef.xml

Font size

plot_fontsize

• Font size used in plots for ticks, etc.
• range in [0, 100]
• default: 12
• views_confdef.xml

Details level

details_level

• Level of details for plotting the parallel coordinates.
• range in [0, 100]
• default: 50
• views_confdef.xml

Max emph lbls

max_emphlbl

• More that this number of entities selected at once will not get individual labels.
• range in [-1, 20]
• default: 5
• views_confdef.xml

Dot shape

dot_shape

• Shape of the dots
• single option in {s, o, p, h, 8, ., ,, *, +, x}
• default: s
• views_confdef.xml

Transparency off

alpha_off

• Turn transparency off.
• yes/no
• default: no
• views_confdef.xml

Add stamp

add_stamp

• Stamp redescription above the figure.
• yes/no
• default: no
• views_confdef.xml

Detailed missing

miss_details

• Show details of different missing values.
• yes/no
• default: yes
• views_confdef.xml

Draw uncovered

draw_delta

• Draw uncovered entities (map)
• yes/no
• default: yes
• views_confdef.xml

Click entities

click_entities

• Show details entities labels on click.
• yes/no
• default: yes
• views_confdef.xml

Hover entities

hover_entities

• Show details entities labels on hover.
• yes/no
• default: no
• views_confdef.xml

Hover coordinates

hover_coords

• Show coordinates on hover.
• yes/no
• default: no
• views_confdef.xml

Show literals contribution

literals_contrib

• Show contribution of literals (PaCo)
• yes/no
• default: no
• views_confdef.xml

Colors: Colors

Custom cluster colors

custom_ccmap

• Custom colors for plotting clusters (matplotlib cmap name, list or dictionnary of hexadecimal colors).
• open text
• default:
• views_confdef.xml

Color LHS

color_l

• Color for plotting the LHS support.
• color #RRGGBB
• default: #fc5864
• views_confdef.xml

Color RHS

color_r

• Color for plotting the RHS support.
• color #RRGGBB
• default: #74a8f6
• views_confdef.xml

Color Intersection

color_i

• Color for plotting the intersection support.
• color #RRGGBB
• default: #662a8d
• views_confdef.xml

Color Uncovered

color_o

• Color for plotting the uncovered entities.
• color #RRGGBB
• default: #e6e6e6
• views_confdef.xml

Color Highlight

color_h

• Color for highlighting entities.
• color #RRGGBB
• default: #ffff00
• views_confdef.xml

List views: Vizualizations

Max nb. clusters

max_clus

• Maximum number of clusters in similarity list view.
• open text
• default: 20
• views_confdef.xml

Values for blocks

blocks_disp_values

• Property to use as value for blocks.
• single option in {fractions, percentages, counts}
• default: fractions
• views_confdef.xml

Re-order r.ids for blocks

blocks_reorder_rids

• Reorder redescriptions for blocks.
• yes/no
• default: yes
• views_confdef.xml

Show r.ids for blocks

blocks_show_rids

• Show redescriptions ids for blocks.
• yes/no
• default: yes
• views_confdef.xml

Show values for blocks

blocks_show_values

• Show values for blocks.
• yes/no
• default: no
• views_confdef.xml

Draw selection block

blocks_show_emph

• Draw special block for selection.
• yes/no
• default: yes
• views_confdef.xml

Support parts

supp_part_clus

• Support parts to use in computing clusters.
• multiple options in {Exx, Exo, Eox, Eoo}
• default: {Exx}
• views_confdef.xml

Missing support parts

miss_part_clus

• Missing support parts to use in computing clusters.
• multiple options in {Exm, Emx, Eom, Emo, Emm}
• default: {}
• views_confdef.xml

Maps elems.: Elements

Map background opacity

map_back_alpha

• Opacity of the background image.
• range in [0, 100]
• default: 100
• views_confdef.xml

Map background scale

map_back_scale

• Scale of the background image.
• range in [0, 100]
• default: 100
• views_confdef.xml

Map background

map_back

• Background to use for the map (only with Basemap, not with Cartopy).
• single option in {none, shaded relief, etopo relief, NASA Blue Marble}
• default: none
• views_confdef.xml

Overlay

map_lines_overlay

• Overlay map lines on top of other plot elements.
• yes/no
• default: no
• views_confdef.xml

Polygons

map_poly

• Use polygons to represent entities on the map
• yes/no
• default: yes
• views_confdef.xml

Map natural areas

map_elem_area

• Map natural areas to paint.
• multiple options in {seas, continents, lakes}
• default: {seas}
• views_confdef.xml

Map natural elements

map_elem_natural

• Map natural elements to draw.
• multiple options in {coasts, rivers}
• default: {coasts}
• views_confdef.xml

Map geopolitic elements

map_elem_geop

• Map geopolitic borders to paint.
• multiple options in {countries, states}
• default: {countries}
• views_confdef.xml

Map imaginary circles

map_elem_circ

• Map imaginary circles to draw.
• multiple options in {parallels, meridians}
• default: {}
• views_confdef.xml

Line color

line_color

• Color for drawing map lines.
• color #RRGGBB
• default: #808080
• views_confdef.xml

Sea color

sea_color

• Color for filling water bodies in maps.
• color #RRGGBB
• default: #f9fcff
• views_confdef.xml

Land color

land_color

• Color for filling land areas in maps.
• color #RRGGBB
• default: #ffffff
• views_confdef.xml

Maps proj.: Projection

Left corner longitude

llon

• Map boundary’s left corner longitude.
• range in [-180.0, 181.0]
• default: 181.0
• views_confdef.xml

Right corner longitude

ulon

• Map boundary’s right corner longitude.
• range in [-180.0, 181.0]
• default: 181.0
• views_confdef.xml

Lower corner latitude

llat

• Map boundary’s lower corner latitude.
• range in [-180.0, 181.0]
• default: 181.0
• views_confdef.xml

Upper corner latitude

ulat

• Map boundary’s upper corner latitude.
• range in [-180.0, 181.0]
• default: 181.0
• views_confdef.xml

Resolution

map_resolution

• Resolution of the map elements.
• single option in {crude, low, intermediate, high, full}
• default: crude
• views_confdef.xml

Projection

map_proj

• Projection to use for the map.
• single option in {none, Plate Carree, Transverse Mercator, Mercator, Lambert Cylindrical, Lambert Conformal, Lambert Azimuthal Equal Area, Miller, Rotated Pole, Gnomonic, Stereographic, North Polar Stereo, South Polar Stereo, Orthographic, Mollweide, Robinson, Interrupted Goode Homolosine, Geostationary, Albers Equal Area, Azimuthal Equidistant, Sinusoidal}
• default: Miller
• views_confdef.xml

Other parameters

It is possible to adjust the parameters through the interface then export them to a file. In particular, a template can be generated and exported which can be used to run the ReReMi mining algorithm from the command line, after filling in input/output parameters, which are specified here.

Further parameters pertain to randomization, cross-fold tests, server parameters, etc.

File Names: Data Input

LHS data filename

LHS_data

• Complete filename for the left hand side data.
• open text
• default:
• inout_confdef.xml

RHS data filename

RHS_data

• Complete filename for the right hand side data.
• open text
• default:
• inout_confdef.xml

Data repertory

data_rep

• Repertory where the data files are stored.
• open text
• default: __TMP_DIR__
• inout_confdef.xml

LHS data basename

data_l

• Basename of the file containing the left hand side data.
• open text
• default: left
• inout_confdef.xml

RHS data basename

data_r

• Basename of the file containing the right hand side data.
• open text
• default: right
• inout_confdef.xml

LHS data extension

ext_l

• Extension of the file containing the left hand side data.
• open text
• default: .csv
• inout_confdef.xml

RHS data extension

ext_r

• Extension of the file containing the right hand side data.
• open text
• default: .csv
• inout_confdef.xml

NA representation

NA_str

• String representation for missing values (CSV format).
• open text
• default: nan
• inout_confdef.xml

Delimiter

delim_in

• Field delimiter (CSV format).
• single option in {(auto), TAB, ;, ,, SPC}
• default: (auto)
• inout_confdef.xml

Time year first

time_yearfirst

• Whether to interpret the first value in an ambiguous 3-integer date as the year (dateutils).
• single option in {(auto), yes, no}
• default: (auto)
• inout_confdef.xml

Time day first

time_dayfirst

• Whether to interpret the first value in an ambiguous 3-integer date as the day (dateutils).
• single option in {(auto), yes, no}
• default: (auto)
• inout_confdef.xml

File Names: Results Output

Results repertory

result_rep

• Repertory where to save the results.
• open text
• default: __TMP_DIR__
• inout_confdef.xml

Results data filename

out_base

• Name of the file where to store the results.
• open text
• default: results
• inout_confdef.xml

Queries extension

ext_queries

• Extension of the file where to save the queries.
• open text
• default: .queries
• inout_confdef.xml

Supports extension

ext_support

• Extension of the file where to save the supports.
• open text
• default: .supports
• inout_confdef.xml

Log data extension

ext_log

• Extension of the file where to save log traces.
• open text
• default: .txt
• inout_confdef.xml

Queries filename

queries_file

• Complete filename for the queries.
• open text
• default:
• inout_confdef.xml

Named queries filename

queries_named_file

• Complete filename for the queries with variable names.
• open text
• default:
• inout_confdef.xml

Support filename

support_file

• Complete filename for the support.
• open text
• default:
• inout_confdef.xml

Advanced: File Names

Series ID

series_id

• Series ID to replace __SID__ in filenames.
• open text
• default:
• inout_confdef.xml

Logfile

logfile

• File to log. + means generate filename from queries output file. - means stdout.
• open text
• default: -
• inout_confdef.xml

Pairs store filename

pairs_store

• Name of the file where to store initial pairs.
• open text
• default:
• inout_confdef.xml

Alternate queries

queries_alternate

• Alternate filename or suffix for the queries, for post-processing.
• open text
• default:
• inout_confdef.xml

Advanced: Extra

Mask for LHS variables

mask_vars_LHS

• Mask for enabling or disabling LHS variables, overriding any info in the data file.
• open text
• default:
• inout_confdef.xml

Mask for RHS variables

mask_vars_RHS

• Mask for enabling or disabling RHS variables, overriding any info in the data file.
• open text
• default:
• inout_confdef.xml

Mask for rows

mask_rows

• Mask for enabling or disabling rows, overriding any info in the data file.
• open text
• default:
• inout_confdef.xml

Extensions repertory

extensions_rep

• Repertory where extension data files are stored.
• open text
• default: __TMP_DIR__
• inout_confdef.xml

Extensions data filenames

extensions_names

• Keys and names of the extension data files.
• open text
• default:
• inout_confdef.xml

Files defining reds fields

fields_rdefs

• Files defining extra redescriptions fields.
• open text
• default:
• inout_confdef.xml

Files defining vars fields

fields_vdefs

• Files defining extra variables fields.
• open text
• default:
• inout_confdef.xml

Files defining reds actions

actions_rdefs

• Files defining extra redescriptions actions.
• open text
• default:
• inout_confdef.xml

Activated extensions

activated_extensions

• Data extensions to activate.
• multiple options in {(auto), geoplus}
• default: {}
• inout_confdef.xml

Network: Work client

Results delay

results_delay

• Delay between results checks by the client, in seconds.
• open text
• default: 5
• network_confdef.xml

Network: Work server

Workserver IP

workserver_ip

• IP adress of the server executing the computational tasks or local for local computations.
• open text
• default: local
• network_confdef.xml

Workserver Port

workserver_port

• Listening port of the server executing the computational tasks.
• open text
• default: 55444
• network_confdef.xml

Workserver Key

workserver_authkey

• Authentification key on the server executing the computational tasks.
• open text
• default: sesame
• network_confdef.xml

Client Id

workserver_clientid

• Client ID to reconnect on the server.
• open text
• default: 0
• network_confdef.xml

Extensions: geoplus

H grid percentile

hgrid_percentile

• Percentile to determine grid height.
• range in [0.0, 100.0]
• default: 90.0
• dataext_confdef.xml

W grid percentile

wgrid_percentile

• Factor to multiply grid width.
• range in [0.0, 100.0]
• default: 90.0
• dataext_confdef.xml

Distance type

dst_type

• Distance to use when computing the grid
• single option in {globe, flat}
• default: globe
• dataext_confdef.xml

Draw tighter borders

after_cut

• Draw tighter border between distant sites.
• yes/no
• default: yes
• dataext_confdef.xml

Folds: Automatic splits

Strips direction

coo_dim

• Coordinate to cut the strips along (-1 and -2 for cut along longitudes and latitudes, respectively).
• open text
• default: -1
• folds_confdef.xml

Granularity

grain

• Granularity of the cut (lower means narrower strips).
• open text
• default: 1.0
• folds_confdef.xml

Variable

folds_col

• Name of the variable to use for splitting the data into subsets.
• open text
• default:
• folds_confdef.xml

Nb folds

nb_folds

• Number of parts the data should be split into.
• range in [2, 20]
• default: 5
• folds_confdef.xml

Randomization: Data Input

Traits data filename

traits_data

• Complete filename for the traits data.
• open text
• default:
• rnd_confdef.xml

Traits data basename

data_t

• Basename of the file containing the traits data.
• open text
• default:
• rnd_confdef.xml

Traits data extension

ext_t

• Extension of the file containing the traits data.
• open text
• default: .csv
• rnd_confdef.xml

Name count variable

count_vname

• Name for the variable that stores occurrence counts.
• open text
• default: COUNTS
• rnd_confdef.xml

Selection redescription

select_red

• Redescription used to select data rows.
• open text
• default:
• rnd_confdef.xml

Agg. precision

agg_prec

• Precision of aggregated data.
• open text
• default: -1
• rnd_confdef.xml

Randomization: Series

Randomization serie size

rnd_series_size

• Number of runs per randomization series.
• open text
• default: 10
• rnd_confdef.xml

Randomization seed

rnd_seed

• Seed for the random number generator.
• open text
• default: -1
• rnd_confdef.xml

Randomization meth.

rnd_meth

• Randomization method used to generate data.
• multiple options in {none, permute_RHS, permute_LHS, permute_traits, shuffle_traits, swaprnd_occs}
• default: {none}
• rnd_confdef.xml