def bpi2012_cancelled():
INP_PATH = "logs/"
EXP_NAME = "bpi2012_cancelled"
LOG_NAME = "BPI_Challenge_2012_tagged_cancelled.xes"
OUTPUTFOLDER = "EnglishBPI/"
results_folder = "bpi2012_cancelled_results"
log_path_seq = "BPI_Challenge_2012_tagged_cancelled_{}.xes"
results_file = "BPI_Challenge_2012_tagged_cancelled_results.txt"
for nr, i in enumerate((5, 15, 25)):
ex = ExperimentRunner(experiment_name=EXP_NAME,
output_file=results_file,
results_folder=results_folder,
inp_path=INP_PATH,
log_name=LOG_NAME,
output_folder=OUTPUTFOLDER,
log_template=log_path_seq,
dt_max_depth=10,
dt_min_leaf=10,
selection_method="coverage",
coverage_threshold=i,
sequence_threshold=5)
with open("train_" + results_file, "a+") as f:
f.write("\n")
with open("test_" + results_file, "a+") as f:
f.write("\n")
#if nr == 0:
# ex.prepare_cross_validation()
# ex.prepare_data()
ex.train_and_eval_benchmark()
def synth_tr1_coverage():
results_folder = "synth_tr"
INP_PATH = "logs/"
LOG_NAME = "synth_tr_tagged.xes"
OUTPUTFOLDER = "Synthetic/"
log_path_seq = "synth_tr_tagged_{}.xes"
results_file = "synth_tr_coverage_results_1.txt"
for i in (5, 15, 25):
ex = ExperimentRunner(output_file=results_file,
results_folder=results_folder,
inp_path=INP_PATH,
log_name=LOG_NAME,
output_folder=OUTPUTFOLDER,
log_template=log_path_seq,
dt_max_depth=10,
dt_min_leaf=5,
selection_method="coverage",
coverage_threshold=i,
experiment_name="tr1_cov")
with open("train_" + results_file, "a+") as f:
f.write("\n")
with open("test_" + results_file, "a+") as f:
f.write("\n")
# ex.prepare_cross_validation()
# ex.prepare_data()
ex.train_and_eval_benchmark()
def sepsis_er_dwd():
INP_PATH = "logs/"
EXP_NAME = "sepsis_er"
LOG_NAME = "sepsis_tagged_er.xes"
OUTPUTFOLDER = "SepsisDWD/"
results_folder = "sepsis_er_dwd_results"
log_path_seq = "sepsis_tagged_er_{}.xes"
results_file = "sepsis_results_data_5.txt"
payload_type = "normal" #dwd, normal or both
payload = True
payload_settings = "sepsis_settings.cfg"
payload_dwd_settings = {
"ignored": [
"Diagnosis", "Diagnose", "time:timestamp", "concept: name",
"Label", "lifecycle: transition"
]
}
reencode = False
for nr, i in enumerate((5, 15, 25)):
ex = ExperimentRunner(experiment_name=EXP_NAME,
output_file=results_file,
results_folder=results_folder,
inp_path=INP_PATH,
log_name=LOG_NAME,
output_folder=OUTPUTFOLDER,
log_template=log_path_seq,
dt_max_depth=5,
dt_min_leaf=5,
selection_method="coverage",
coverage_threshold=i,
sequence_threshold=5,
payload_type=payload_type,
payload=payload,
payload_settings=payload_settings,
reencode=reencode,
payload_dwd_settings=payload_dwd_settings)
with open("train_" + results_file, "a+") as f:
f.write("\n")
with open("test_" + results_file, "a+") as f:
f.write("\n")
#if nr == 0:
# #ex.prepare_cross_validation()
# ex.prepare_data()
ex.train_and_eval_benchmark()
def bpi2011_dwd_data_m16():
INP_PATH = "logs/"
EXP_NAME = "test"
LOG_NAME = "EnglishBPIChallenge2011_tagged_m16_prep.xes"
OUTPUTFOLDER = "EnglishBPI/"
results_folder = "bpi2011_t101_results"
log_path_seq = "EnglishBPIChallenge2011_tagged_m16_prep_{}.xes"
results_file = "bpi2011_t101_results.txt"
payload = True
payload_settings = "bpi2011_settings.cfg"
selection_method = "coverage"
payload_type = "normal" #dwd, normal or both
ignored = [
"time:timestamp", "concept: name", "Label", "Start date", "End date",
"Diagnosis", "Diagnosis code", "Diagnosis Treatment", "Combination ID",
"Treatment code", "Activity code"
]
payload_dwd_settings = {"ignored": ignored}
for nr, i in enumerate((5, 15, 25)):
ex = ExperimentRunner(experiment_name=EXP_NAME,
output_file=results_file,
results_folder=results_folder,
inp_path=INP_PATH,
log_name=LOG_NAME,
output_folder=OUTPUTFOLDER,
log_template=log_path_seq,
dt_max_depth=10,
dt_min_leaf=5,
selection_method=selection_method,
coverage_threshold=i,
sequence_threshold=5,
payload=payload,
payload_settings=payload_settings,
payload_type=payload_type,
payload_dwd_settings=payload_dwd_settings)
with open("train_" + results_file, "a+") as f:
f.write("\n")
with open("test_" + results_file, "a+") as f:
f.write("\n")
#if nr == 0:
#ex.prepare_cross_validation()
#ex.prepare_data()
ex.train_and_eval_benchmark()
def sepsis_er_data():
INP_PATH = "logs/"
EXP_NAME = "sepsis_er"
LOG_NAME = "sepsis_tagged_er.xes"
OUTPUTFOLDER = "Sepsis/"
results_folder = "sepsis_er_results"
log_path_seq = "sepsis_tagged_er_{}.xes"
results_file = "sepsis_er_nodata.txt"
payload = True
payload_settings = "sepsis_settings.cfg"
reencode = False
for nr, i in enumerate((5, 15, 25)):
ex = ExperimentRunner(experiment_name=EXP_NAME,
output_file=results_file,
results_folder=results_folder,
inp_path=INP_PATH,
log_name=LOG_NAME,
output_folder=OUTPUTFOLDER,
log_template=log_path_seq,
dt_max_depth=10,
dt_min_leaf=5,
selection_method="coverage",
coverage_threshold=i,
sequence_threshold=5,
payload=payload,
payload_settings=payload_settings,
reencode=reencode)
with open("train_" + results_file, "a+") as f:
f.write("\n")
with open("test_" + results_file, "a+") as f:
f.write("\n")
if nr == 0:
ex.prepare_cross_validation()
ex.prepare_data()
ex.train_and_eval_benchmark()
ex.clean_data()
def sig_disc_test():
INP_PATH = "logs/"
EXP_NAME = "test"
LOG_NAME = "multi.xes"
OUTPUTFOLDER = "Sepsis/"
results_folder = "multi_results"
log_path_seq = "multi_{}.xes"
results_file = "multi_results.txt"
for nr, i in enumerate((5, 15, 25)):
ex = ExperimentRunner(experiment_name=EXP_NAME,
output_file=results_file,
results_folder=results_folder,
inp_path=INP_PATH,
log_name=LOG_NAME,
output_folder=OUTPUTFOLDER,
log_template=log_path_seq,
dt_max_depth=10,
dt_min_leaf=10,
selection_method="coverage",
coverage_threshold=i,
sequence_threshold=5)
with open("train_" + results_file, "a+") as f:
f.write("\n")
with open("test_" + results_file, "a+") as f:
f.write("\n")
if nr == 0:
#ex.prepare_cross_validation()
ex.prepare_data()
ex.train_and_eval_benchmark()
def bpi2011_data_cc():
INP_PATH = "logs/"
EXP_NAME = "test"
LOG_NAME = "EnglishBPIChallenge2011_tagged_cc_prep.xes"
OUTPUTFOLDER = "EnglishBPI/"
results_folder = "bpi2011_cc_data_prep_res"
log_path_seq = "EnglishBPIChallenge2011_tagged_cc_prep_{}.xes"
results_file = "bpi2011_cc_data_prep_res.txt"
payload = True
payload_settings = "bpi2011_settings.cfg"
selection_method = "coverage"
#selection_method = "rf_importance"
#selection_method = "none"
for nr, i in enumerate((5, 15, 25)):
ex = ExperimentRunner(experiment_name=EXP_NAME,
output_file=results_file,
results_folder=results_folder,
inp_path=INP_PATH,
log_name=LOG_NAME,
output_folder=OUTPUTFOLDER,
log_template=log_path_seq,
dt_max_depth=10,
dt_min_leaf=5,
selection_method=selection_method,
coverage_threshold=i,
sequence_threshold=5,
payload=payload,
payload_settings=payload_settings)
with open("train_" + results_file, "a+") as f:
f.write("\n")
with open("test_" + results_file, "a+") as f:
f.write("\n")
#if nr == 0:
# ex.prepare_cross_validation()
# ex.prepare_data()
ex.train_and_eval_benchmark()
def synth_tr1():
results_folder = "synth_tr"
INP_PATH = "logs/"
LOG_NAME = "synth_tr_tagged.xes"
OUTPUTFOLDER = "Synthetic/"
log_path_seq = "synth_tr_tagged_{}.xes"
results_file = "synth_tr_results_1.txt"
ex = ExperimentRunner(output_file=results_file,
results_folder=results_folder,
inp_path=INP_PATH,
log_name=LOG_NAME,
output_folder=OUTPUTFOLDER,
log_template=log_path_seq,
dt_min_leaf=5,
dt_max_depth=10,
experiment_name="tr1")
# ex.prepare_cross_validation()
# ex.prepare_data()
ex.train_and_eval_benchmark()
def output_pos_neg_test():
INP_PATH = "logs/"
EXP_NAME = "output_pos_neg_data"
LOG_NAME = "output_pos_and_neg.xes"
OUTPUTFOLDER = "payload/"
results_folder = "output_pos_neg_res"
log_path_seq = "output_pos_and_neg_{}.xes"
results_file = "output_pos_neg.txt"
payload = True
payload_settings = "output_pos_and_neg_settings.cfg"
for nr, i in enumerate((5, 15, 25)):
ex = ExperimentRunner(experiment_name=EXP_NAME,
output_file=results_file,
results_folder=results_folder,
inp_path=INP_PATH,
log_name=LOG_NAME,
output_folder=OUTPUTFOLDER,
log_template=log_path_seq,
dt_max_depth=10,
dt_min_leaf=10,
selection_method="coverage",
coverage_threshold=i,
sequence_threshold=5,
payload=payload,
payload_settings=payload_settings)
with open("train_" + results_file, "a+") as f:
f.write("\n")
with open("test_" + results_file, "a+") as f:
f.write("\n")
#if nr == 0:
# ex.prepare_cross_validation()
# ex.prepare_data()
ex.train_and_eval_benchmark()
def synth_mr_new():
results_folder = "synth_mra_results"
EXP_NAME = "xray"
payload = False
selection_method = "coverage"
INP_PATH = "logs/"
LOG_NAME = "synth_mr_tagged.xes"
OUTPUTFOLDER = "Synthetic/"
log_path_seq = "synth_mr_tagged_{}.xes"
results_file = "synth_coverage_mr_results_1.txt"
for nr, i in enumerate((5, 15, 25)):
ex = ExperimentRunner(experiment_name=EXP_NAME,
output_file=results_file,
results_folder=results_folder,
inp_path=INP_PATH,
log_name=LOG_NAME,
output_folder=OUTPUTFOLDER,
payload=payload,
log_template=log_path_seq,
dt_max_depth=10,
dt_min_leaf=5,
selection_method=selection_method,
coverage_threshold=i,
sequence_threshold=5)
with open("train_" + results_file, "a+") as f:
f.write("\n")
with open("test_" + results_file, "a+") as f:
f.write("\n")
if nr == 0:
ex.prepare_cross_validation()
ex.prepare_data()
ex.train_and_eval_benchmark()
parser = argparse.ArgumentParser(description='Process some integers.')
parser.add_argument('--start',
metavar='start',
type=int,
help='start',
default=0)
parser.add_argument('--end',
metavar='end',
type=int,
help='end',
default=1011)
parser.add_argument('--timeout',
metavar='timeout',
type=float,
help='timeout',
default=5)
args = parser.parse_args()
start = args.start
end = args.end
timeout = args.timeout
print("timeout={}, start={}, end={}".format(timeout, start, end))
rulesGenerator = ConstraintRulesGenerator()
rulesDict = rulesGenerator.singleCummulativeForEachConstraint()
runner = ExperimentRunner()
before = time.time()
runner.run("SingleCummulativeForEachConstraint", rulesDict, timeout, start, end)
print("time={} seconds".format(time.time()-before))