def test_save_task():
tasks.save_task({"description": "Do something", "status": "1"})
task_list = tasks.get_task_list()
assert type(task_list) is list
found = False
for task in task_list:
if task["description"] == "Do something":
found = True
assert found
def test_delete_task():
task_id = tasks.save_task({
"description": "this is a delete task",
"status": "1"
})
task_list = tasks.get_task_list()
found = False
for task in task_list:
if "delete" in task['description']:
found = True
assert found
tasks.delete_task(task_id)
task_list = tasks.get_task_list()
found = False
for task in task_list:
if "delete" in task['description']:
found = True
assert not found
def get(self):
with locks.global_lock:
session_id = self.get_cookie('session_id')
session = auth.load_session(session_id)
if session is None:
self.redirect('/login')
return
team_list = list(team.get_all_teams())
task_list = list(tasks.get_task_list())
self.write(
render_template('scoreboard.html',
session=session,
team_list=team_list,
task_list=task_list))
def main():
if '--help' in sys.argv:
print('Usage: {} [--force]'.format(sys.argv[0]))
print('Generates all tasks for all teams in advance')
print()
print('Options:')
print(' --force Regenerate even if generated tasks are up-to-date')
sys.exit(1)
generate_func = task_gen.generate if '--force' in sys.argv else task_gen.maybe_generate
team_list = list(team.get_all_teams())
task_list = list(tasks.get_task_list())
for tm in team_list:
for ts in task_list:
token = task_gen.get_token(team_name=tm.team_name,
task_id=ts.task_id)
generate_func(task_id=ts.task_id, token=token, team=tm)
def get(self):
with locks.global_lock:
session = auth.load_session(self.get_cookie('session_id'))
submissions = team.get_all_submissions()
if session is None or not session.is_admin:
self.write(
render_template('admin_error.html',
error=lc.get('not_admin')))
return
self.write(
render_template(
'admin.html',
session=session,
tasks=list(tasks.get_task_list()),
submissions=list(submissions),
teams=list(team.get_all_teams()),
groups=dict(tasks.get_group_dict()),
read_task=tasks.read_task,
build_group_path=tasks.build_group_path,
task_gen=task_gen,
))
def get_argparser():
parser = argparse.ArgumentParser()
# Performance Options
parser.add_argument("--local_rank", type=int, default=0)
parser.add_argument("--random_seed",
type=int,
default=42,
help="random seed (default: 42)")
parser.add_argument("--num_workers",
type=int,
default=1,
help='number of workers (default: 1)')
# Datset Options
parser.add_argument("--data_root",
type=str,
default='data',
help="path to Dataset")
parser.add_argument("--dataset",
type=str,
default='voc',
choices=['voc', 'ade'],
help='Name of dataset')
parser.add_argument("--num_classes",
type=int,
default=None,
help="num classes (default: None)")
# Method Options
# BE CAREFUL USING THIS, THEY WILL OVERRIDE ALL THE OTHER PARAMETERS.
parser.add_argument(
"--method",
type=str,
default=None,
choices=['FT', 'LWF', 'LWF-MC', 'ILT', 'EWC', 'RW', 'PI', 'MiB'],
help=
"The method you want to use. BE CAREFUL USING THIS, IT MAY OVERRIDE OTHER PARAMETERS."
)
# Train Options
parser.add_argument("--epochs",
type=int,
default=30,
help="epoch number (default: 30)")
parser.add_argument(
"--fix_bn",
action='store_true',
default=False,
help='fix batch normalization during training (default: False)')
parser.add_argument("--batch_size",
type=int,
default=4,
help='batch size (default: 4)')
parser.add_argument("--crop_size",
type=int,
default=512,
help="crop size (default: 513)")
parser.add_argument("--lr",
type=float,
default=0.007,
help="learning rate (default: 0.007)")
parser.add_argument("--momentum",
type=float,
default=0.9,
help='momentum for SGD (default: 0.9)')
parser.add_argument("--weight_decay",
type=float,
default=1e-4,
help='weight decay (default: 1e-4)')
parser.add_argument("--lr_policy",
type=str,
default='poly',
choices=['poly', 'step'],
help="lr schedule policy (default: poly)")
parser.add_argument("--lr_decay_step",
type=int,
default=5000,
help="decay step for stepLR (default: 5000)")
parser.add_argument("--lr_decay_factor",
type=float,
default=0.1,
help="decay factor for stepLR (default: 0.1)")
parser.add_argument("--lr_power",
type=float,
default=0.9,
help="power for polyLR (default: 0.9)")
parser.add_argument("--bce",
default=False,
action='store_true',
help="Whether to use BCE or not (default: no)")
# Validation Options
parser.add_argument("--val_on_trainset",
action='store_true',
default=False,
help="enable validation on train set (default: False)")
parser.add_argument(
"--cross_val",
action='store_true',
default=False,
help="If validate on training or on validation (default: Train)")
parser.add_argument("--crop_val",
action='store_false',
default=True,
help='do crop for validation (default: True)')
# Logging Options
parser.add_argument("--logdir",
type=str,
default='./logs',
help="path to Log directory (default: ./logs)")
parser.add_argument(
"--name",
type=str,
default='Experiment',
help=
"name of the experiment - to append to log directory (default: Experiment)"
)
parser.add_argument(
"--sample_num",
type=int,
default=0,
help='number of samples for visualization (default: 0)')
parser.add_argument("--debug",
action='store_true',
default=False,
help="verbose option")
parser.add_argument("--visualize",
action='store_false',
default=True,
help="visualization on tensorboard (def: Yes)")
parser.add_argument("--print_interval",
type=int,
default=10,
help="print interval of loss (default: 10)")
parser.add_argument("--val_interval",
type=int,
default=1,
help="epoch interval for eval (default: 1)")
parser.add_argument("--ckpt_interval",
type=int,
default=1,
help="epoch interval for saving model (default: 1)")
# Model Options
parser.add_argument("--backbone",
type=str,
default='resnet101',
choices=['resnet50', 'resnet101'],
help='backbone for the body (def: resnet50)')
parser.add_argument("--output_stride",
type=int,
default=16,
choices=[8, 16],
help='stride for the backbone (def: 16)')
parser.add_argument("--no_pretrained",
action='store_true',
default=False,
help='Wheather to use pretrained or not (def: True)')
parser.add_argument("--norm_act",
type=str,
default="iabn_sync",
choices=['iabn_sync', 'iabn', 'abn', 'std'],
help='Which BN to use (def: abn_sync')
parser.add_argument(
"--fusion-mode",
metavar="NAME",
type=str,
choices=["mean", "voting", "max"],
default="mean",
help="How to fuse the outputs. Options: 'mean', 'voting', 'max'")
parser.add_argument(
"--pooling",
type=int,
default=32,
help='pooling in ASPP for the validation phase (def: 32)')
# Test and Checkpoint options
parser.add_argument(
"--test",
action='store_true',
default=False,
help="Whether to train or test only (def: train and test)")
parser.add_argument(
"--ckpt",
default=None,
type=str,
help=
"path to trained model. Leave it None if you want to retrain your model"
)
# Parameters for Knowledge Distillation of ILTSS (https://arxiv.org/abs/1907.13372)
parser.add_argument(
"--freeze",
action='store_true',
default=False,
help="Use this to freeze the feature extractor in incremental steps")
parser.add_argument(
"--loss_de",
type=float,
default=0., # Distillation on Encoder
help="Set this hyperparameter to a value greater than "
"0 to enable distillation on Encoder (L2)")
parser.add_argument(
"--loss_kd",
type=float,
default=0., # Distillation on Output
help="Set this hyperparameter to a value greater than "
"0 to enable Knowlesge Distillation (Soft-CrossEntropy)")
# Parameters for EWC, RW, and SI (from Riemannian Walks https://arxiv.org/abs/1801.10112)
parser.add_argument("--regularizer",
default=None,
type=str,
choices=['ewc', 'rw', 'pi'],
help="regularizer you want to use. Default is None")
parser.add_argument(
"--reg_importance",
type=float,
default=1.,
help="set this par to a value greater than 0 to enable regularization")
parser.add_argument(
"--reg_alpha",
type=float,
default=0.9,
help=
"Hyperparameter for RW and EWC that controls the update of Fisher Matrix"
)
parser.add_argument("--reg_no_normalize",
action='store_true',
default=False,
help="If EWC, RW, PI must be normalized or not")
parser.add_argument(
"--reg_iterations",
type=int,
default=10,
help=
"If RW, the number of iterations after each the update of the score is done"
)
# Arguments for ICaRL (from https://arxiv.org/abs/1611.07725)
parser.add_argument("--icarl",
default=False,
action='store_true',
help="If enable ICaRL or not (def is not)")
parser.add_argument(
"--icarl_importance",
type=float,
default=1.,
help="the regularization importance in ICaRL (def is 1.)")
parser.add_argument(
"--icarl_disjoint",
action='store_true',
default=False,
help="Which version of icarl is to use (def: combined)")
parser.add_argument("--icarl_bkg",
action='store_true',
default=False,
help="If use background from GT (def: No)")
# METHODS
parser.add_argument(
"--init_balanced",
default=False,
action='store_true',
help="Enable Background-based initialization for new classes")
parser.add_argument(
"--unkd",
default=False,
action='store_true',
help=
"Enable Unbiased Knowledge Distillation instead of Knowledge Distillation"
)
parser.add_argument(
"--alpha",
default=1.,
type=float,
help="The parameter to hard-ify the soft-labels. Def is 1.")
parser.add_argument(
"--unce",
default=False,
action='store_true',
help="Enable Unbiased Cross Entropy instead of CrossEntropy")
# Incremental parameters
parser.add_argument("--task",
type=str,
default="19-1",
choices=tasks.get_task_list(),
help="Task to be executed (default: 19-1)")
parser.add_argument("--step",
type=int,
default=0,
help="The incremental step in execution (default: 0)")
parser.add_argument(
"--no_mask",
action='store_true',
default=False,
help="Use this to not mask the old classes in new training set")
parser.add_argument(
"--overlap",
action='store_true',
default=False,
help="Use this to not use the new classes in the old training set")
parser.add_argument(
"--step_ckpt",
default=None,
type=str,
help=
"path to trained model at previous step. Leave it None if you want to use def path"
)
parser.add_argument('--opt_level',
type=str,
choices=['O0', 'O1', 'O2', 'O3'],
default='O0')
return parser
def get_generated_task_list(team_name):
for task in tasks.get_task_list():
token = get_token(team_name=team_name, task_id=task.task_id)
yield get_generated_task(task_id=task.task_id,
token=token,
team=team.read_team(team_name))
def get_task(task_id):
task = get_task_list(task_id)
if task == None:
abort(404)
return jsonify({'task': task})
def test_get_tasks():
task_l = tasks.get_task_list()
assert type(task_l) is list
def reports():
task_list=tasks.get_task_list(tasks.get_task_id_list())
report_format_list = tasks.get_report_formats()
return render_template('reports.html', tasks=task_list, reports=report_format_list)
sys.path.append(parentdir)
from scan import scan, is_requested, is_running, get_newprogress, check_for_logging
from questions import getQuestions, submitAnswers
import tasks
from setup import set_dhcp, set_static_ip, get_ip, get_subnet
from update_code import update_code
from configparser import ConfigParser
import re
from zipfile import ZipFile
import json
import time
app = Flask(__name__)
IpAddressen = []
task_list=tasks.get_task_list(tasks.get_task_id_list())
report_format_list = tasks.get_report_formats()
errorList = []
progr=0
task_id_for_progr = " "
temp_deviceName= " "
already_running=False
thread_list=[]
active_hosts=[]
conf_id = "698f691e-7489-11df-9d8c-002264764cea"
def create_dailylog():
todays_logs = datetime.now().strftime("%d-%m-%Y")
# check if the directory for the logs exists and if the logfile for this day exists
