def __init__(self, settings, log):
"""constructor"""
self.space = self.define_search_space(settings)
# Read inputs
self.log = log
self._split_timeline(0.8, settings['read_options']['one_timestamp'])
self.org_log = copy.deepcopy(log)
self.org_log_train = copy.deepcopy(self.log_train)
self.org_log_valdn = copy.deepcopy(self.log_valdn)
# Load settings
self.settings = settings
self.temp_output = os.path.join('output_files', sup.folder_id())
if not os.path.exists(self.temp_output):
os.makedirs(self.temp_output)
self.file_name = os.path.join(self.temp_output,
sup.file_id(prefix='OP_'))
# Results file
if not os.path.exists(self.file_name):
open(self.file_name, 'w').close()
# Trials object to track progress
self.bayes_trials = Trials()
self.best_output = None
self.best_parms = dict()
self.best_similarity = 0
def __init__(self, parms, log, ac_index, rl_index):
"""constructor"""
self.space = self.define_search_space(parms)
self.log = copy.deepcopy(log)
self.ac_index = ac_index
self.rl_index = rl_index
# Load settings
self.parms = parms
self.temp_output = parms['output']
if not os.path.exists(self.temp_output):
os.makedirs(self.temp_output)
os.makedirs(os.path.join(self.temp_output, 'opt_parms'))
self.file_name = sup.file_id(prefix='OP_')
# Results file
if not os.path.exists(os.path.join(self.temp_output, self.file_name)):
open(os.path.join(self.temp_output, self.file_name), 'w').close()
self.conn = {
'partition': 'main',
'mem': str(32000),
'cpus': str(10),
'env': 'deep_generator_pip',
'script': os.path.join('model_training', 'slurm_trainer.py')
}
self.slurm_workers = 50
self.best_output = None
self.best_parms = dict()
self.best_loss = 1
def __init__(self, parms, log_train, log_valdn, ac_index, ac_weights):
"""constructor"""
self.space = self.define_search_space(parms)
self.log_train = copy.deepcopy(log_train)
self.log_valdn = copy.deepcopy(log_valdn)
self.ac_index = ac_index
self.ac_weights = ac_weights
# Load settings
self.parms = parms
self.temp_output = parms['output']
if not os.path.exists(self.temp_output):
os.makedirs(self.temp_output)
os.makedirs(os.path.join(self.temp_output, 'opt_parms'))
self.file_name = os.path.join(self.temp_output, sup.file_id(prefix='OP_'))
# Results file
if not os.path.exists(self.file_name):
open(self.file_name, 'w').close()
self.conn = {'partition': 'main',
'mem': str(32000),
'cpus': str(10),
'env': 'deep_sim3',
'script': os.path.join('core_modules',
'times_allocator',
'slurm_trainer.py')}
self.slurm_workers = 50
self.best_output = None
self.best_parms = dict()
self.best_loss = 1
def _execute_simulator(bimp_path, temp_path, model):
"""Executes BIMP Simulations.
Args:
settings (dict): Path to jar and file names
rep (int): repetition number
"""
sim_log = os.path.join(temp_path, sup.file_id('SIM_'))
args = ['java', '-jar', bimp_path, model, '-csv', sim_log]
subprocess.run(args, check=True, stdout=subprocess.PIPE)
return sim_log
def _export_results(self, output_path) -> None:
# Save results
pd.DataFrame(self.sim_values).to_csv(os.path.join(
output_path, sup.file_id(prefix='SE_')),
index=False)
# Save logs
log_test = self.log_test[~self.log_test.task.isin(['Start', 'End'])]
log_test.to_csv(os.path.join(
output_path,
'tst_' + self.parms['gl']['file'].split('.')[0] + '.csv'),
index=False)
if self.parms['gl']['save_models']:
paths = [
'bpmn_models', 'embedded_path', 'ia_gen_path',
'seq_flow_gen_path', 'times_gen_path'
]
sources = list()
for path in paths:
for root, dirs, files in os.walk(self.parms['gl'][path]):
for file in files:
if self.parms['gl']['file'].split('.')[0] in file:
sources.append(os.path.join(root, file))
for source in sources:
base_folder = os.path.join(
output_path, os.path.basename(os.path.dirname(source)))
if not os.path.exists(base_folder):
os.makedirs(base_folder)
destination = os.path.join(base_folder,
os.path.basename(source))
# Copy dl models
allowed_ext = self._define_model_path({
**self.parms['gl'],
**self.parms['t_gen']
})
is_dual = self.parms['t_gen']['model_type'] == 'dual_inter'
if is_dual and ('times_gen_models' in source) and any(
[x in source for x in allowed_ext]):
shutil.copyfile(source, destination)
elif not is_dual and ('times_gen_models' in source) and any([
self.parms['gl']['file'].split('.')[0] + x in source
for x in allowed_ext
]):
shutil.copyfile(source, destination)
# copy other models
folders = ['bpmn_models', 'embedded_matix', 'ia_gen_models']
allowed_ext = [
'.emb', '.bpmn', '_mpdf.json', '_prf.json',
'_prf_meta.json', '_mpdf_meta.json', '_meta.json'
]
if any([x in source for x in folders]) and any([
self.parms['gl']['file'].split('.')[0] + x in source
for x in allowed_ext
]):
shutil.copyfile(source, destination)
def _export_results(self, output_path) -> None:
# Save results
pd.DataFrame(self.sim_values).to_csv(
os.path.join(self.output_route, sup.file_id(prefix='SE_')),
index=False)
# Save logs
log_test = self.log[~self.log.task.isin(['Start', 'End'])]
log_test.to_csv(
os.path.join(self.output_route, 'tst_'+
self.parms['model_file'].split('.')[0]+'.csv'),
index=False)
def export_params(self):
configs_files = list()
for config in self.space:
config['ac_index'] = self.ac_index
config['rl_index'] = self.rl_index
conf_file = sup.file_id(prefix='CNF_', extension='.json')
sup.create_json(
config, os.path.join(self.temp_output, 'opt_parms', conf_file))
configs_files.append(conf_file)
self.log.to_csv(os.path.join(self.temp_output, 'opt_parms',
'train.csv'),
index=False,
encoding='utf-8')
return configs_files
def __init__(self, parms, log_train, log_valdn, ac_index, ac_weights):
"""constructor"""
self.space = self.define_search_space(parms)
self.log_train = copy.deepcopy(log_train)
self.log_valdn = copy.deepcopy(log_valdn)
self.ac_index = ac_index
self.ac_weights = ac_weights
# Load settings
self.parms = parms
self.temp_output = parms['output']
if not os.path.exists(self.temp_output):
os.makedirs(self.temp_output)
self.file_name = os.path.join(self.temp_output,
sup.file_id(prefix='OP_'))
# Results file
if not os.path.exists(self.file_name):
open(self.file_name, 'w').close()
# Trials object to track progress
self.bayes_trials = Trials()
self.best_output = None
self.best_parms = dict()
self.best_loss = 1
