def clean_exit(reason=None):
print('Exiting gracefully (reason: {0})'.format(reason), file=sys.stderr)
# revert all changes to iptables
helpers.run(
'iptables-save | grep -v "searchExamServer" | iptables-restore -w')
helpers.run(
'iptables-legacy-save | grep -v "searchExamServer" | iptables-legacy-restore -w'
)
# if isDeb9OrNewer():
# url_whitelist = helpers.run('diff --unchanged-group-format="" <(echo "^{p}://{h}" | sed "s/\./\\\./g") {f}'.format(f = urlWhitelistFile, p = gladosProto, h = gladosHost))
# else:
# url_whitelist = helpers.run('diff --unchanged-group-format="" <(echo "^{p}://{h}") {f}'.format(f = urlWhitelistFile, p = gladosProto, h = gladosHost))
helpers.run('umount /run/initramfs/newroot 2>/dev/null')
helpers.run('umount -l /run/initramfs/{base,exam} 2>/dev/null')
if isDeb9OrNewer(): helpers.run('squid -k reconfigure') # iptables stays
exit(0)
def compute(tpl):
"""Takes a combination of gamma and nu value and trains and
evaluates a OC-SVM using the current feature set"""
g = tpl[0]
n = tpl[1]
ocsvm = svm.OneClassSVM(nu=n, kernel="rbf", gamma=g, cache_size=1000)
errorRate = func.calcErrorRate(func.run(ocsvm, config["folds"], trS, teS))
tmpResults[errorRate] = (g, n, lookUp[subsetCnt])
print str(format(errorRate[0], '.20f')) + " " + str(
format(errorRate[1], '.20f')) + " : gamma= " + str(g) + " nu= " + str(
n) + " features= " + str(lookUp[subsetCnt])
else:
util.cli = "riff"
if args.push_secret is None or len(args.push_secret) <= 0:
util.push_secret = ""
else:
if args.image_prefix is None or len(args.image_prefix) <= 0:
raise Exception("An --image-prefix must be provided when using --push-secret")
util.push_secret = args.push_secret
util.image_prefix = args.image_prefix
if args.pull_secret is None or len(args.pull_secret) <= 0:
util.pull_secret = ""
else:
util.pull_secret = args.pull_secret
if args.manifest is None or len(args.manifest) <= 0:
if args.pfs:
raise Exception("A manifest must be provided for PFS")
util.manifest = "stable"
else:
util.manifest = args.manifest
import setup, teardown, functions, eventing
setup.run()
functions.run()
eventing.run()
teardown.run()
elapsed_time = time.time() - start_time
elapsed_min = int(elapsed_time / 60)
elapsed_sec = int(elapsed_time - (elapsed_min * 60))
print("DONE in {m} min {s} sec".format(m=elapsed_min, s=elapsed_sec))
]
skplt.estimators.plot_feature_importances(extra_clf, max_num_features=5)
selected_x_set = selected_x.transform(X_data)
selected_x_set_new = selected_x.transform(data_new)
selected_x_set_old = selected_x.transform(data_oldies)
# %%
models_old_sig = {}
models_new_sig = {}
models_whole_sig = {}
models_whole = {}
models_old_s = {}
models_new_s = {}
models_whole_s = {}
models_old_sig = func.run(data_oldies_sig, y_old)
models_new_sig = func.run(data_new_sig, y_new)
models_whole_sig = func.run(X_data_sig, y_sig)
models_whole = func.run(X_data, y)
models_old_s = func.run(selected_x_set_old, y_old)
models_new_s = func.run(selected_x_set_new, y_new)
models_whole_s = func.run(selected_x_set, y)
test = func.run_ann(X_data, y)
models_old_sig["ann_basic"] = func.run_ann(data_oldies_sig, y_old)
models_new_sig["ann_basic"] = func.run_ann(data_new_sig, y_new)
models_whole_sig["ann_basic"] = func.run_ann(X_data_sig, y_sig)
models_whole["ann_basic"] = func.run_ann(X_data, y)
models_old_s["ann_basic"] = func.run_ann(selected_x_set_old, y_old)
models_new_s["ann_basic"] = func.run_ann(selected_x_set_new, y_new)
from classes import Pause
from functions import get_driver, run
from sys import exit
if __name__ == '__main__':
sleep_duration = Pause()
counter = 0
omegle = "https://www.omegle.com/"
print(f"[*] Navigating to {omegle}")
while True:
driver = get_driver(omegle)
driver = run(driver, sleep_duration)
driver.get_screenshot_as_file(f'./screen_shots/run_{counter}.png')
print(f"[*] Iteration {counter} complete | Image: run_{counter}.png")
counter += 1
driver.quit()
sleep_duration.reset()
from functions import run
semestre = '2021-1'
ramos = [
# Pregrado
# {'nrc': 15218, 'sigla': 'ICS3313', 'sec': 3},
# {'nrc': 10207, 'sigla': 'IHI2315', 'sec': 1},
# {'nrc': 10725, 'sigla': 'IIC2133', 'sec': 1},
{'nrc': 10732, 'sigla': 'IIC2143', 'sec': 1},
# Práctica II
# {'nrc': 24183, 'sigla': 'ING2005', 'sec': 1},
# {'nrc': 11844, 'sigla': 'ING2001', 'sec': 1},
# Magíster
{'nrc': 18871, 'sigla': 'IIC3697', 'sec': 1},
{'nrc': 10930, 'sigla': 'IIC3724', 'sec': 1},
{'nrc': 11856, 'sigla': 'ING4921', 'sec': 1},
]
if __name__ == '__main__':
run(semestre, ramos, interval=60)
#! /usr/bin/python3 from functions import run run()
import argparse
import importlib
import os
import sys
from functions import create_exp_name_and_datetime_path, merge_cfg_with_cli, run
from human_id import generate_id
if __name__ == "__main__":
# parser = argparse.ArgumentParser()
exp = sys.argv[1].split("/")[-1].split(".")[0]
module = importlib.import_module("." + exp, package="experiments")
Experiment = getattr(module, "Experiment")
cfg = getattr(module, "Config")
parser = merge_cfg_with_cli(cfg)
parser.add_argument("exp")
args = parser.parse_args()
path = create_exp_name_and_datetime_path(Experiment)
path = os.path.join("results", path)
args.run_id = generate_id()
run(Experiment, args, path)
import functions as hard hard.run()
def compute(tpl):
global counter
global threshold
g = tpl[0]
n = tpl[1]
counter = counter + 1
ocsvm = svm.OneClassSVM(nu=n, kernel="rbf", gamma=g, cache_size=1000)
errorRate = func.calcErrorRate(func.run(ocsvm, config["folds"], trS, teS))
tmpResults[errorRate] = (g, n, lookUp[subsetCnt])
print str(format(errorRate[0], '.20f')) + " " + str(format(errorRate[1], '.20f')) + " : gamma= " + str(g) + " nu= " + str(n) + " features= " + str(lookUp[subsetCnt])
if counter == rejected:
threshold = findBestErrorRate(results)
t = open('data/results/stopping-rule/result-' + str(config["folds"]) + '.txt', 'a+')
sys.stdout = t
print "Results for feature subset / model parameters for " + str(config["folds"]) + "-folded CV with " + str(
len(gammaVal)) + \
" gamma values and " + str(len(nuVal)) + " nu values:\n"
print "Stopping rule threshold:"
print "gamma :" + str(threshold[1][0])
print "nu :" + str(threshold[1][1])
print "feature subset :" + str(threshold[1][2])
print "threshold : %s%% false alarm rate, %s%% miss rate" % (
str(threshold[0][0] * 100), str(threshold[0][1] * 100)) + "\n"
sys.stdout = temp
t.close()
elif counter > rejected:
t = open('data/results/stopping-rule/result-' + str(config["folds"]) + '.txt', 'a+')
sys.stdout = t
if (threshold[0][0] > errorRate[0]) or (threshold[0][0] == errorRate[0] and threshold[0][1] > errorRate[1]):
print "Best result found:"
print "gamma :" + str(g)
print "nu :" + str(n)
print "feature subset :" + str(lookUp[subsetCnt])
print "grid search results : %s%% false alarm rate, %s%% miss rate" % (
str(errorRate[0] * 100), str(errorRate[1] * 100))
print "\n"
print "Started on " + started.strftime("%Y-%m-%d %H:%M")
print "Finished on " + datetime.datetime.now().strftime("%Y-%m-%d %H:%M")
print "Execution time " + str(datetime.datetime.now() - started)
print "------------------------------------------------------------"
temp.write("\n" + "Started on " + started.strftime("%Y-%m-%d %H:%M") + "\n")
temp.write("Finished on " + datetime.datetime.now().strftime("%Y-%m-%d %H:%M") + "\n")
temp.write("------------------------------------------------------------\n")
sys.exit("Stopping rule")
elif counter == candidates:
print "Best result found (not better than the threshold):"
print "gamma :" + str(g)
print "nu :" + str(n)
print "feature subset :" + str(lookUp[subsetCnt])
print "grid search results : %s%% false alarm rate, %s%% miss rate" % (
str(errorRate[0] * 100), str(errorRate[1] * 100))
print "\n"
print "Started on " + started.strftime("%Y-%m-%d %H:%M")
print "Finished on " + datetime.datetime.now().strftime("%Y-%m-%d %H:%M")
print "Execution time " + str(datetime.datetime.now() - started)
print "------------------------------------------------------------"
sys.stdout = temp
t.close()
pool.apply_async(run.run,
args=(days, horizon, SoC, NL, b, num_samples, taus,
inpEndo, inpExo, tar, lambdas, perfectFC))
for b in bo
]
my_res = [p.get() for p in my_res]
pool.close()
print("--- %s seconds ---" % (time.time() - start_time))
# Run just one process during testing
#taus = np.arange(0.1,0.91,0.1)
#fn.run(days,horizon,SoC,NL,False,num_samples,taus,inpEndo,inpExo,tar,lambdas,perfectFC)
# Run the uncontrolled charging
#fn.uncontrolledCharging(perfectFC,SoC,lambdas)
'''
# RUNNING DETERMINISTIC WITH PERFECT FORECASTS
# NOTE THAT THE FUNCTION RUN SHOULD BE MANUALLY ADJUSTED FOR THIS SPECIAL CASE
#num_samples = 1
taus = np.arange(0.1,0.91,0.1)
fn.run(days,horizon,SoC,NL,False,num_samples,taus,inpEndo,inpExo,tar,lambdas,perfectFC)
'''
'''
# Train the GBRT models
# Use the same hyperparameters for all forecast horizons.
all_files = glob.glob(os.path.join("CrossValidation", "*.txt"))
params = pd.concat((pd.read_csv(f, sep="\t") for f in all_files), ignore_index=True).sort_values(by=['CRPS']).iloc[0,0:6]
if __name__ == '__main__':
pool = mp.Pool(processes=64)
horizon = np.arange(1,97,1)
my_res = [pool.apply_async(fc_fn.gbrt_training, args=(h,inpEndo,inpExo,tar,params)) for h in horizon]
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
try:
s.bind(('localhost', 18493))
except OSError:
print('Program is already running, exitting.', file=sys.stderr)
sys.exit(1)
signal.signal(signal.SIGINT, trap)
signal.signal(signal.SIGTERM, trap)
# import time
# time.sleep(10)
# allow bonjour/zeroconf
helpers.run(
'iptables -I INPUT -p udp --dport 5353 -d 224.0.0.251 -m comment --comment "searchExamServer" -j ACCEPT'
)
helpers.run(
'iptables -I OUTPUT -p udp --dport 5353 -d 224.0.0.251 -m comment --comment "searchExamServer" -j ACCEPT'
)
display = helpers.get_env("DISPLAY")
xauthority = helpers.get_env("XAUTHORITY")
env = {
'DISPLAY': display,
'XAUTHORITY': xauthority,
'LANG': helpers.get_env('LANG')
}
# create the directory structure and cleanup
for path in [
