NiaARMTS

class niaarmts.NiaARMTS.NiaARMTS(dimension, lower, upper, features, transactions, interval, alpha, beta, gamma, delta)

Bases: Problem

add_rule_to_archive(full_rule, antecedent, consequent, fitness, start, end, support, confidence, inclusion, amplitude)

Add the rule to the archive if its fitness is greater than zero and it’s not already present.

Parameters:
  • full_rule (list) – The full rule generated from the solution.

  • antecedent (list) – The antecedent part of the rule.

  • consequent (list) – The consequent part of the rule.

  • fitness (float) – The fitness value of the rule.

  • start (timestamp) – The start timestamp for the rule.

  • end (timestamp) – The end timestamp for the rule.

  • support (float) – Support value for the rule.

  • confidence (float) – Confidence value for the rule.

  • inclusion (float) – Inclusion metric for the rule.

  • amplitude (float) – Amplitude metric for the rule.

cut_point(sol, num_attr)

Calculate cut point based on the solution and the number of attributes.

get_rule_archive()

Return the archive of all valid rules (those with fitness > 0), sorted by fitness in descending order.

rule_representation(rule)

Generate a string representation of a rule for easier comparison and to avoid duplicates. :param rule: The rule to represent as a string. :type rule: list

Returns:

A string representation of the rule.

Return type:

str

save_rules_to_csv(file_path)

Save the archived rules to a CSV file, sorted by fitness (descending).

Parameters:

file_path (str) – The path to save the CSV file.

save_rules_to_json(file_path)

Save the archived rules to a JSON file, sorted by fitness (descending).

Parameters:

file_path (str) – The path to save the JSON file.