Running configuration

This page details how to use the Cleora run command to define the embedding resources at runtime.

Synopsis

cleora [options 2] <params 1> [options 2] <params 2> …

Run options

• input

Using input param: –input or -i

Param description: A parameter that defines path for input file. You can use also absolute path or relative path.

• file type

Using file type param: –type or -t

Param description: This parameter is responsible for defining the input file extension to the algorithm. Cleora supports two kinds of input files .tsv (tab-separated values) and .json.

• dimension

Using dimension param: –dimenstion or -d

Param description: Embedding dimension size.

• number of iterations

Using number of iterations param: –dimenstion or -d

param Description: Set maximum number of iterations.

• columns

Using columnns param: –columns or -c

Param description: Set column names (max. 12), with modifiers from list: [transient::, reflexive::, complex::]

Modifiers	Description
transient	The field is virtual - it is considered during embedding process, no entity is written for the column
reflexive	The field is composite, containing multiple entity identifiers separated by space in TSV or an array in JSON
complex	The field is reflexive, which means that it interacts with itself, additional output file is written for every such field
ignore	The field is ignored, no output file is written for the field

Allowed combinations of modifiers are:

• transient

• complex

• transient::complex

• reflexive::complex

For TSV datasets containing composite fields (categorical array), multiple items within a field are then separated by space.

The specification of an input format is as follows:

--columns="[column modifiers, ::]<column_name> ..."

Combinations which don’t make sense are:

Modifiers	Description
reflexive	This would represent an identity relation
transient::reflexive	This would generate no output
reflexive::transient::complex	This would generate no output

Picture below representation how works column modifiers:

• relation name

Using relation param: –relation-name or -r

Param description: Name of the relation, for output filename generation.

• prepend field name

Using prepend field name param: –relation-name or -r

Param description: Prameter that responsible for prepending field name to entity in output.

• log every n

Using log every n params : –log-every-n or -l

Param description: Set log output for every N lines

• in memory embedding calculation

Using log in memory embedding calculation param: –in-memory-embedding-calculation or -e

Param description: Parameter that responsible for using calculate embeddings in memory or with memory-mapped files. Default is on (setting -e 0). If you want off use -e 1.

-output dir

Using output dir param: –output-dir or -o

Param description: Set output directory for files with embeddings.

-output format

Using output format param: –output-format or -o

Param Description: A parameter that defines the format of the output file. Possible output format are textfile (.txt) and numpy (.npy)

Examples Cleora run configuration

Remember before you will first run cleora training (after download binary file from repository) to set execute file permission using chmod +x

chmod +x cleora
./cleora -i files/samples/edgelist_sample.tsv
         --columns="complex::reflexive::a b complex::c"
         -d 128
         -n 5
         --relation-name=test_realation_name
         -p 0