distances Package

distances Package

distances Module

class ABXpy.distances.distances.Features_Accessor(times, features)[source]

Bases: object

get_features_from_raw(items)[source]
get_features_from_splitted(items)[source]
ABXpy.distances.distances.compute_distances(feature_file, feature_group, pair_file, distance_file, distance, normalized, n_cpu=None, mem=1000, feature_file_as_list=False)[source]
ABXpy.distances.distances.create_distance_jobs(pair_file, distance_file, n_cpu, buffer_max_size=100)[source]

Divide the work load into smaller blocks to be passed to the cpus

Parameters
pair_file: string

hdf5 task file, or at least an hdf5 file containing a ‘unique_pairs’ dataset

distance_file: string

hdf5 file taht will contain the distance datasets

n_cpu:

number of cpus tu use

block_ceil_size: int

maximum size in RAM of a block in Mb

ABXpy.distances.distances.run_distance_job(job_description, distance_file, distance, feature_files, feature_groups, splitted_features, job_id, normalize, distance_file_lock=None)[source]
ABXpy.distances.distances.worker(args)[source]