def __init__(self,
dir,
authorIds={},
context=None,
start=True,
baseDir=None):
if not os.path.isdir(dir):
os.mkdir(dir)
if not list(os.listdir(dir)):
self.ix = create_in(dir, Index.schema)
if not baseDir:
baseDir = os.path.join(os.path.split(dir)[0])
self.ix = whoosh.index.open_dir(dir)
self.searchers = []
self.failedDir = os.path.join(baseDir, "failed")
utils.ensureDir(self.failedDir)
self.incomingDir = os.path.join(baseDir, "incoming")
utils.ensureDir(self.incomingDir)
self.indexer = threading.Thread(target=Index.indexLoop, args=[self])
self.logger = open(os.path.join(baseDir, "index.log"), "a")
self.stopping = False
if start:
self.startIndexer()
self.counts = {}
def start(self):
global mongod_port
global mongod
if self.proc:
print >> sys.stderr, "probable bug: self.proc already set in start()"
return
self.ensure_test_dirs()
dir_name = smoke_db_prefix + "/data/db/sconsTests/"
self.port = int(mongod_port)
self.slave = False
if 'slave' in self.kwargs:
dir_name = smoke_db_prefix + '/data/db/sconsTestsSlave/'
srcport = mongod_port
self.port += 1
self.slave = True
if os.path.exists(dir_name):
if 'slave' in self.kwargs:
argv = [utils.find_python(), "buildscripts/cleanbb.py", '--nokill', dir_name]
else:
argv = [utils.find_python(), "buildscripts/cleanbb.py", dir_name]
call(argv)
utils.ensureDir(dir_name)
argv = [mongod_executable, "--port", str(self.port), "--dbpath", dir_name]
# This should always be set for tests
argv += ['--setParameter', 'enableTestCommands=1']
if self.kwargs.get('small_oplog'):
argv += ["--master", "--oplogSize", "511"]
if self.kwargs.get('small_oplog_rs'):
argv += ["--replSet", "foo", "--oplogSize", "511"]
if self.slave:
argv += ['--slave', '--source', 'localhost:' + str(srcport)]
if self.kwargs.get('no_journal'):
argv += ['--nojournal']
if self.kwargs.get('no_preallocj'):
argv += ['--nopreallocj']
if self.kwargs.get('auth'):
argv += ['--auth']
self.auth = True
if self.kwargs.get('use_ssl'):
argv += ['--sslOnNormalPorts',
'--sslPEMKeyFile', 'jstests/libs/server.pem',
'--sslCAFile', 'jstests/libs/ca.pem']
print "running " + " ".join(argv)
self.proc = self._start(buildlogger(argv, is_global=True))
if not self.did_mongod_start(self.port):
raise Exception("Failed to start mongod")
if self.auth:
self.setup_admin_user(self.port)
if self.slave:
local = Connection(port=self.port, slave_okay=True).local
synced = False
while not synced:
synced = True
for source in local.sources.find(fields=["syncedTo"]):
synced = synced and "syncedTo" in source and source["syncedTo"]
def run_finer_lrs(init_param='kaiming', device='cpu'):
dist_grid = [ExampleDistribution()
] + [RadialDataDistribution(d=2**k) for k in range(7)]
std_grid = [0.1, 0.5, 1.0, 2.0]
# bi_grid = [('zero', 0.0), ('he+5', 0.0), ('he+1', 0.0), ('kink_uniform', 0.0)] \
# + [(bim, big) for big in std_grid for bim in ['normal', 'uniform']] \
# + [('pos-unif', 1.0), ('neg-unif', 1.0), ('kink-unif', 1.0), ('kink-neg-unif', 1.0),
# ('kink-neg-point', 0.0)]
bi_grid = [('zero', 0.0), ('unif', 1.0), ('unif-pos', 1.0),
('unif-neg', 1.0), ('kink-neg-unif', 1.0), ('pytorch', 1.0),
('kink-neg-point', 0.0)]
for opt in ['gd', 'gd-mom', 'adam']:
for dist in dist_grid:
d = dist.get_x_dim()
for bim, big in bi_grid:
folder_name = f'{init_param}_{opt}_{dist.get_name()}_{bim}-{big:g}'
path = Path(custom_paths.get_results_path()
) / 'nn_comparison' / folder_name
best_lr_file = Path(custom_paths.get_results_path(
)) / 'nn_comparison' / f'{folder_name}_bestlr.pkl'
if not utils.existsFile(best_lr_file):
sys.stderr.write(
'best lr file {best_lr_file} does not exist!\n')
continue
best_lr = utils.deserialize(best_lr_file)
lr_grid = [best_lr * (2**(k / 8)) for k in range(-3, 4)]
for lr in lr_grid:
print(f'Running combination {folder_name} with lr {lr:g}')
file = path / f'{lr:g}.pkl'
utils.ensureDir(file)
if utils.existsFile(file):
continue
n_rep = 2 if d == 64 else 1
trainer = SimpleParallelTrainer(n_parallel=100 // n_rep,
n_train=256 * d,
n_valid=1024,
n_test=1024,
data_distribution=dist,
lr=lr,
bias_init_gain=big,
batch_size=256,
bias_init_mode=bim,
init_param=init_param,
n_epochs=8192 // d,
seed=0,
device=device,
n_hidden=512,
opt=opt,
valid_epoch_interval=64 //
d,
n_rep=n_rep)
results = trainer.fit(do_plot=False, verbose=False)
if results is None:
print('Got NaN values')
utils.serialize(file, {
'trainer': trainer,
'results': results
})
def execute_mc(offset_str, base_dir='./mc-data/'):
param_combinations = get_param_combinations()
num_processes = max(1, multiprocessing.cpu_count() // 2)
pool = multiprocessing.Pool(processes=num_processes)
utils.ensureDir(base_dir + 'mc-data-{}/'.format(offset_str))
pool.map(KerasMCRunner(offset_str, base_dir),
param_combinations,
chunksize=1)
pool.terminate()
pool.join()
def start(self):
global mongod_port
global mongod
if self.proc:
print >> sys.stderr, "probable bug: self.proc already set in start()"
return
self.ensure_test_dirs()
dir_name = smoke_db_prefix + "/data/db/sconsTests/"
self.port = int(mongod_port)
self.slave = False
if 'slave' in self.kwargs:
dir_name = smoke_db_prefix + '/data/db/sconsTestsSlave/'
srcport = mongod_port
self.port += 1
self.slave = True
if os.path.exists(dir_name):
if 'slave' in self.kwargs:
argv = [utils.find_python(), "buildscripts/cleanbb.py", '--nokill', dir_name]
else:
argv = [utils.find_python(), "buildscripts/cleanbb.py", dir_name]
call(argv)
utils.ensureDir(dir_name)
argv = [mongod_executable, "--port", str(self.port), "--dbpath", dir_name]
if self.kwargs.get('small_oplog'):
argv += ["--master", "--oplogSize", "511"]
if self.kwargs.get('small_oplog_rs'):
argv += ["--replSet", "foo", "--oplogSize", "511"]
if self.slave:
argv += ['--slave', '--source', 'localhost:' + str(srcport)]
if self.kwargs.get('no_journal'):
argv += ['--nojournal']
if self.kwargs.get('no_preallocj'):
argv += ['--nopreallocj']
if self.kwargs.get('auth'):
argv += ['--auth']
self.auth = True
print "running " + " ".join(argv)
self.proc = self._start(buildlogger(argv, is_global=True))
if not self.did_mongod_start(self.port):
raise Exception("Failed to start mongod")
if self.auth:
self.setup_admin_user(self.port)
if self.slave:
local = Connection(port=self.port, slave_okay=True).local
synced = False
while not synced:
synced = True
for source in local.sources.find(fields=["syncedTo"]):
synced = synced and "syncedTo" in source and source["syncedTo"]
def start(self):
global mongod_port
global mongod
if self.proc:
print >> sys.stderr, "probable bug: self.proc already set in start()"
return
self.ensure_test_dirs()
dir_name = smoke_db_prefix + "/data/db/sconsTests/"
self.port = int(mongod_port)
self.slave = False
if "slave" in self.kwargs:
dir_name = smoke_db_prefix + "/data/db/sconsTestsSlave/"
srcport = mongod_port
self.port += 1
self.slave = True
if os.path.exists(dir_name):
if "slave" in self.kwargs:
argv = [utils.find_python(), "buildscripts/cleanbb.py", "--nokill", dir_name]
else:
argv = [utils.find_python(), "buildscripts/cleanbb.py", dir_name]
call(argv)
utils.ensureDir(dir_name)
argv = [mongod_executable, "--port", str(self.port), "--dbpath", dir_name]
if self.kwargs.get("small_oplog"):
argv += ["--master", "--oplogSize", "511"]
if self.kwargs.get("small_oplog_rs"):
argv += ["--replSet", "foo", "--oplogSize", "511"]
if self.slave:
argv += ["--slave", "--source", "localhost:" + str(srcport)]
if self.kwargs.get("no_journal"):
argv += ["--nojournal"]
if self.kwargs.get("no_preallocj"):
argv += ["--nopreallocj"]
if self.kwargs.get("auth"):
argv += ["--auth"]
self.auth = True
print "running " + " ".join(argv)
self.proc = self._start(buildlogger(argv, is_global=True))
if not self.did_mongod_start(self.port):
raise Exception("Failed to start mongod")
if self.auth:
self.setup_admin_user(self.port)
if self.slave:
local = Connection(port=self.port, slave_okay=True).local
synced = False
while not synced:
synced = True
for source in local.sources.find(fields=["syncedTo"]):
synced = synced and "syncedTo" in source and source["syncedTo"]
def execute_mc(offset_str,
base_dir='./mc-data/',
use_sgd=False,
use_early_stopping=False,
use_sufficient_stopping=True,
use_small_lr=False,
initialize_custom=False):
# creates a thread pool and runs all computation tasks
param_combinations = get_param_combinations()
num_processes = max(1, multiprocessing.cpu_count() // 2)
pool = multiprocessing.Pool(processes=num_processes)
utils.ensureDir(base_dir + 'mc-data-{}/'.format(offset_str))
pool.map(OffsetMCRunner(offset_str, base_dir, use_sgd, use_early_stopping,
use_sufficient_stopping, use_small_lr,
initialize_custom),
param_combinations,
chunksize=1)
pool.terminate()
pool.join()
def run_old(init_param='kaiming', device='cpu'):
dist_grid = [ExampleDistribution()
] + [RBFDataDistribution(d=2**k) for k in range(7)]
std_grid = [0.1, 0.5, 1.0, 2.0]
bi_grid = [('zero', 0.0), ('he+5', 0.0), ('he+1', 0.0), ('kink_uniform', 0.0)] \
+ [(bim, big) for big in std_grid for bim in ['normal', 'uniform']]
for opt in ['gd', 'gd-mom', 'adam']:
base_lr = 1e-2 if opt == 'adam' else (
4e-1 if init_param == 'ntk' else 8e-3)
lr_grid = [base_lr * np.sqrt(2)**k for k in range(-8, 9)]
for dist in dist_grid:
for bim, big in bi_grid:
folder_name = f'{init_param}_{opt}_{dist.get_name()}_{bim}-{big:g}'
path = Path(custom_paths.get_results_path()
) / 'nn_comparison' / folder_name
for lr in lr_grid:
print(f'Running combination {folder_name} with lr {lr:g}')
file = path / f'{lr:g}.pkl'
utils.ensureDir(file)
if utils.existsFile(file):
continue
torch.cuda.empty_cache()
trainer = SimpleParallelTrainer(n_parallel=100,
n_train=256,
n_valid=1024,
n_test=1024,
data_distribution=dist,
lr=lr,
bias_init_gain=big,
bias_init_mode=bim,
init_param=init_param,
n_epochs=10000,
seed=0,
device=device,
n_hidden=256,
opt=opt)
results = trainer.fit(do_plot=False, verbose=False)
if results is None:
print('Got NaN values')
utils.serialize(file, {
'trainer': trainer,
'results': results
})
def start(self):
global mongod_port
global mongod
if self.proc:
print >> sys.stderr, "probable bug: self.proc already set in start()"
return
self.ensure_test_dirs()
dir_name = smoke_db_prefix + "/data/db/sconsTests/"
self.port = int(mongod_port)
self.slave = False
if 'slave' in self.kwargs:
dir_name = smoke_db_prefix + '/data/db/sconsTestsSlave/'
srcport = mongod_port
self.port += 1
self.slave = True
if os.path.exists(dir_name):
if 'slave' in self.kwargs:
argv = [
utils.find_python(), "buildscripts/cleanbb.py", '--nokill',
dir_name
]
else:
argv = [
utils.find_python(), "buildscripts/cleanbb.py", dir_name
]
call(argv)
utils.ensureDir(dir_name)
argv = [
mongod_executable, "--port",
str(self.port), "--dbpath", dir_name
]
# These parameters are alwas set for tests
# SERVER-9137 Added httpinterface parameter to keep previous behavior
argv += ['--setParameter', 'enableTestCommands=1', '--httpinterface']
if self.kwargs.get('small_oplog'):
argv += ["--master", "--oplogSize", "511"]
params = self.kwargs.get('set_parameters', None)
if params:
for p in params.split(','):
argv += ['--setParameter', p]
if self.kwargs.get('small_oplog_rs'):
argv += ["--replSet", "foo", "--oplogSize", "511"]
if self.slave:
argv += ['--slave', '--source', 'localhost:' + str(srcport)]
if self.kwargs.get('no_journal'):
argv += ['--nojournal']
if self.kwargs.get('no_preallocj'):
argv += ['--nopreallocj']
if self.kwargs.get('auth'):
argv += ['--auth']
authMechanism = self.kwargs.get('authMechanism', 'MONGODB-CR')
if authMechanism != 'MONGODB-CR':
argv += [
'--setParameter',
'authenticationMechanisms=' + authMechanism
]
self.auth = True
if self.kwargs.get('use_ssl'):
argv += [
'--sslOnNormalPorts', '--sslPEMKeyFile',
'jstests/libs/server.pem', '--sslCAFile',
'jstests/libs/ca.pem', '--sslWeakCertificateValidation'
]
print "running " + " ".join(argv)
self.proc = self._start(buildlogger(argv, is_global=True))
if not self.did_mongod_start(self.port):
raise Exception("Failed to start mongod")
if self.auth:
self.setup_admin_user(self.port)
if self.slave:
local = Connection(port=self.port, slave_okay=True).local
synced = False
while not synced:
synced = True
for source in local.sources.find(fields=["syncedTo"]):
synced = synced and "syncedTo" in source and source[
"syncedTo"]
def ensure_test_dirs(self):
utils.ensureDir(smoke_db_prefix + "/tmp/unittest/")
utils.ensureDir(smoke_db_prefix + "/data/")
utils.ensureDir(smoke_db_prefix + "/data/db/")
def start(self):
global mongod_port
global mongod
if self.proc:
print >> sys.stderr, "probable bug: self.proc already set in start()"
return
self.ensure_test_dirs()
dir_name = smoke_db_prefix + "/data/db/sconsTests/"
self.port = int(mongod_port)
self.slave = False
if "slave" in self.kwargs:
dir_name = smoke_db_prefix + "/data/db/sconsTestsSlave/"
srcport = mongod_port
self.port += 1
self.slave = True
if os.path.exists(dir_name):
if "slave" in self.kwargs:
argv = [utils.find_python(), "buildscripts/cleanbb.py", "--nokill", dir_name]
else:
argv = [utils.find_python(), "buildscripts/cleanbb.py", dir_name]
call(argv)
utils.ensureDir(dir_name)
argv = [mongod_executable, "--port", str(self.port), "--dbpath", dir_name]
# This should always be set for tests
argv += ["--setParameter", "enableTestCommands=1"]
if self.kwargs.get("small_oplog"):
argv += ["--master", "--oplogSize", "511"]
if self.kwargs.get("small_oplog_rs"):
argv += ["--replSet", "foo", "--oplogSize", "511"]
if self.slave:
argv += ["--slave", "--source", "localhost:" + str(srcport)]
if self.kwargs.get("no_journal"):
argv += ["--nojournal"]
if self.kwargs.get("no_preallocj"):
argv += ["--nopreallocj"]
if self.kwargs.get("auth"):
argv += ["--auth"]
authMechanism = self.kwargs.get("authMechanism", "MONGO-CR")
if authMechanism != "MONGO-CR":
argv.append("--setParameter=authenticationMechanisms=" + authMechanism)
self.auth = True
if self.kwargs.get("use_ssl"):
argv += [
"--sslOnNormalPorts",
"--sslPEMKeyFile",
"jstests/libs/server.pem",
"--sslCAFile",
"jstests/libs/ca.pem",
"--sslWeakCertificateValidation",
]
print "running " + " ".join(argv)
self.proc = self._start(buildlogger(argv, is_global=True))
if not self.did_mongod_start(self.port):
raise Exception("Failed to start mongod")
if self.auth:
self.setup_admin_user(self.port)
if self.slave:
local = Connection(port=self.port, slave_okay=True).local
synced = False
while not synced:
synced = True
for source in local.sources.find(fields=["syncedTo"]):
synced = synced and "syncedTo" in source and source["syncedTo"]
def ensure_test_dirs(self):
utils.ensureDir(smoke_db_prefix + "/tmp/unittest/")
utils.ensureDir(smoke_db_prefix + "/data/")
utils.ensureDir(smoke_db_prefix + "/data/db/")
def plot_loss():
n_hidden = 16
n_parallel = 1
lr = 1e-3
n_steps = 201
n_epochs_per_step = 100
np.random.seed(4)
x, y = mc_training.get_standard_dataset()
#y += 0.5
(a, b, c, w) = CheckingNN.createRandomWeights(n_parallel, n_hidden)
x_weights = 1. / len(x) * np.ones(len(x))
train_setups = [TrainingSetup(x, x_weights, y) for i in range(n_parallel)]
lrs = np.array([lr for i in range(n_parallel)])
net = CheckingNN(initial_weights=(a, b, c, w),
train_setups=train_setups,
lrs=lrs)
vars = net.create_training_vars()
losses = []
y_eval = []
x_eval = []
for step in range(n_steps):
pred_diff = net.predict(x, 0) - y
loss = np.dot(pred_diff, pred_diff) / (2 * len(x))
losses.append(loss - 1.0)
x_eval_current = np.sort(
np.hstack([[-3.0, 3.0], -net.b[0, 0, :] / net.a[0, 0, :]]))
x_eval.append(x_eval_current)
y_eval.append(net.predict(x_eval_current, 0))
for step_epoch in range(n_epochs_per_step):
net.train_one_epoch(vars)
plt.figure('Loss', figsize=(3, 2))
plt.semilogy(np.arange(n_steps) * n_epochs_per_step, losses, 'k')
plt.xlabel(r'Epoch $k$')
plt.ylabel(r"$L_D(W_k) - \inf_{k'} L_D(W_{k'})$")
utils.ensureDir('./plots/')
plt.tight_layout()
plt.savefig('./plots/loss.pdf')
plt.figure('NN training', figsize=(6, 4))
linewidth = 0.5
plt.plot(x, y, 'k.')
plt.plot(x_eval[0], y_eval[0], 'k--', linewidth=linewidth, label='Initial')
plt.plot(x_eval[10],
y_eval[10],
'#AAAAAA',
linewidth=linewidth,
label='1000 Epochs')
plt.plot(x_eval[n_steps - 1],
y_eval[n_steps - 1],
'k',
linewidth=linewidth,
label='20000 Epochs')
plt.scatter(x_eval[n_steps - 1][1:-1], y_eval[n_steps - 1][1:-1], s=1)
plt.xlabel(r'$x$')
plt.ylabel(r'$y$')
plt.legend()
plt.tight_layout()
plt.savefig('./plots/training.pdf')
plt.figure('Kinks', figsize=(3, 2))
kink_movement = np.array([x_eval[i][1:-1] for i in range(len(x_eval))])
for i in range(kink_movement.shape[1]):
plt.plot(np.arange(kink_movement.shape[0]) * n_epochs_per_step,
kink_movement[:, i],
'k',
linewidth=linewidth)
plt.xlabel(r'Epoch $k$')
plt.ylabel(r'$-b_{i, k}/a_{i, k}$')
plt.tight_layout()
plt.savefig('./plots/kink_movement.pdf')
def start(self):
global mongod_port
global mongod
if self.proc:
print >> sys.stderr, "probable bug: self.proc already set in start()"
return
self.ensure_test_dirs()
dir_name = smoke_db_prefix + "/data/db/sconsTests/"
self.port = int(mongod_port)
self.slave = False
if 'slave' in self.kwargs:
dir_name = smoke_db_prefix + '/data/db/sconsTestsSlave/'
srcport = mongod_port
self.port += 1
self.slave = True
if os.path.exists(dir_name):
if 'slave' in self.kwargs:
argv = [utils.find_python(), "buildscripts/cleanbb.py", '--nokill', dir_name]
else:
argv = [utils.find_python(), "buildscripts/cleanbb.py", dir_name]
call(argv)
utils.ensureDir(dir_name)
argv = [mongod_executable, "--port", str(self.port), "--dbpath", dir_name]
# These parameters are alwas set for tests
# SERVER-9137 Added httpinterface parameter to keep previous behavior
argv += ['--setParameter', 'enableTestCommands=1', '--httpinterface']
if self.kwargs.get('small_oplog'):
argv += ["--master", "--oplogSize", "511"]
params = self.kwargs.get('set_parameters', None)
if params:
for p in params.split(','): argv += ['--setParameter', p]
if self.kwargs.get('small_oplog_rs'):
argv += ["--replSet", "foo", "--oplogSize", "511"]
if self.slave:
argv += ['--slave', '--source', 'localhost:' + str(srcport)]
if self.kwargs.get('no_journal'):
argv += ['--nojournal']
if self.kwargs.get('no_preallocj'):
argv += ['--nopreallocj']
if self.kwargs.get('auth'):
argv += ['--auth']
authMechanism = self.kwargs.get('authMechanism', 'MONGODB-CR')
if authMechanism != 'MONGODB-CR':
argv += ['--setParameter', 'authenticationMechanisms=' + authMechanism]
self.auth = True
if self.kwargs.get('use_ssl') or self.kwargs.get('use_x509'):
argv += ['--sslMode', "sslOnly",
'--sslPEMKeyFile', 'jstests/libs/server.pem',
'--sslCAFile', 'jstests/libs/ca.pem',
'--sslWeakCertificateValidation']
if self.kwargs.get('use_x509'):
argv += ['--clusterAuthMode','x509'];
self.auth = True
print "running " + " ".join(argv)
self.proc = self._start(buildlogger(argv, is_global=True))
if not self.did_mongod_start(self.port):
raise Exception("Failed to start mongod")
if self.auth:
self.setup_admin_user(self.port)
if self.slave:
local = Connection(port=self.port, slave_okay=True).local
synced = False
while not synced:
synced = True
for source in local.sources.find(fields=["syncedTo"]):
synced = synced and "syncedTo" in source and source["syncedTo"]
def start(self):
global mongod_port
global mongod
global shell_executable
global _debug
global valgrind
global drd
if self.proc:
print >> sys.stderr, "probable bug: self.proc already set in start()"
return
self.ensure_test_dirs()
dir_name = smoke_db_prefix + "/data/db/sconsTests/"
self.port = int(mongod_port)
self.slave = False
if 'slave' in self.kwargs:
dir_name = smoke_db_prefix + '/data/db/sconsTestsSlave/'
srcport = mongod_port
self.port += 1
self.slave = True
if os.path.exists(dir_name):
pass
""" Cleanbb is the most irritating script ever created.
if 'slave' in self.kwargs:
argv = [utils.find_python(), "buildscripts/cleanbb.py", '--nokill', dir_name]
else:
argv = [utils.find_python(), "buildscripts/cleanbb.py", dir_name]
if quiet:
f = open(server_log_file, "a")
try:
call(argv, stdout=f)
finally:
f.close()
else:
call(argv)
"""
utils.ensureDir(dir_name)
argv = [
mongod_executable, "--port",
str(self.port), "--dbpath", dir_name
]
# This should always be set for tests
argv += ['--setParameter', 'enableTestCommands=1']
if valgrind or drd:
argv += ['--setParameter', 'numCachetableBucketMutexes=32']
if self.kwargs.get('small_oplog'):
argv += ["--master", "--oplogSize", "511"]
if self.kwargs.get('small_oplog_rs'):
argv += ["--replSet", "foo", "--oplogSize", "511"]
if self.slave:
argv += ['--slave', '--source', 'localhost:' + str(srcport)]
if self.kwargs.get('no_journal'):
argv += ['--nojournal']
if self.kwargs.get('no_preallocj'):
argv += ['--nopreallocj']
if self.kwargs.get('auth'):
argv += ['--auth']
authMechanism = self.kwargs.get('authMechanism', 'MONGODB-CR')
if authMechanism != 'MONGODB-CR':
argv.append('--setParameter=authenticationMechanisms=' +
authMechanism)
self.auth = True
if len(server_log_file) > 0:
argv += ['--logpath', server_log_file]
if len(smoke_server_opts) > 0:
argv += [smoke_server_opts]
if self.kwargs.get('use_ssl'):
argv += [
'--sslOnNormalPorts', '--sslPEMKeyFile',
'jstests/libs/server.pem', '--sslCAFile',
'jstests/libs/ca.pem', '--sslWeakCertificateValidation'
]
if not quiet:
print "running " + " ".join(argv)
self.proc = self._start(buildlogger(argv, is_global=True))
if not self.did_mongod_start(self.port):
raise Exception("Failed to start mongod")
run_mongo_cmd = [shell_executable, '--port', str(self.port), '--quiet']
if self.kwargs.get('use_ssl'):
run_mongo_cmd += [
'--ssl', '--sslPEMKeyFile', 'jstests/libs/server.pem',
'--sslCAFile', 'jstests/libs/ca.pem'
]
if "true" == check_output(
run_mongo_cmd +
['--eval', 'print(db.runCommand("buildInfo").debug)']).strip():
_debug = True
if self.auth:
self.setup_admin_user(self.port)
if self.slave:
local = Connection(port=self.port, slave_okay=True).local
synced = False
while not synced:
synced = True
for source in local.sources.find(fields=["syncedTo"]):
synced = synced and "syncedTo" in source and source[
"syncedTo"]
def start(self):
global mongod_port
global mongod
global shell_executable
global _debug
global valgrind
global drd
if self.proc:
print >> sys.stderr, "probable bug: self.proc already set in start()"
return
self.ensure_test_dirs()
dir_name = smoke_db_prefix + "/data/db/sconsTests/"
self.port = int(mongod_port)
self.slave = False
if 'slave' in self.kwargs:
dir_name = smoke_db_prefix + '/data/db/sconsTestsSlave/'
srcport = mongod_port
self.port += 1
self.slave = True
if os.path.exists(dir_name):
pass
""" Cleanbb is the most irritating script ever created.
if 'slave' in self.kwargs:
argv = [utils.find_python(), "buildscripts/cleanbb.py", '--nokill', dir_name]
else:
argv = [utils.find_python(), "buildscripts/cleanbb.py", dir_name]
if quiet:
f = open(server_log_file, "a")
try:
call(argv, stdout=f)
finally:
f.close()
else:
call(argv)
"""
utils.ensureDir(dir_name)
argv = [mongod_executable, "--port", str(self.port), "--dbpath", dir_name]
# This should always be set for tests
argv += ['--setParameter', 'enableTestCommands=1']
if valgrind or drd:
argv += ['--setParameter', 'numCachetableBucketMutexes=32']
if self.kwargs.get('small_oplog'):
argv += ["--master", "--oplogSize", "511"]
if self.kwargs.get('small_oplog_rs'):
argv += ["--replSet", "foo", "--oplogSize", "511"]
if self.slave:
argv += ['--slave', '--source', 'localhost:' + str(srcport)]
if self.kwargs.get('no_journal'):
argv += ['--nojournal']
if self.kwargs.get('no_preallocj'):
argv += ['--nopreallocj']
if self.kwargs.get('auth'):
argv += ['--auth']
authMechanism = self.kwargs.get('authMechanism', 'MONGODB-CR')
if authMechanism != 'MONGODB-CR':
argv.append('--setParameter=authenticationMechanisms=' + authMechanism)
self.auth = True
if len(server_log_file) > 0:
argv += ['--logpath', server_log_file]
if len(smoke_server_opts) > 0:
argv += [smoke_server_opts]
if self.kwargs.get('use_ssl'):
argv += ['--sslOnNormalPorts',
'--sslPEMKeyFile', 'jstests/libs/server.pem',
'--sslCAFile', 'jstests/libs/ca.pem',
'--sslWeakCertificateValidation']
if not quiet:
print "running " + " ".join(argv)
self.proc = self._start(buildlogger(argv, is_global=True))
if not self.did_mongod_start(self.port):
raise Exception("Failed to start mongod")
run_mongo_cmd = [shell_executable, '--port', str(self.port), '--quiet']
if self.kwargs.get('use_ssl'):
run_mongo_cmd += ['--ssl',
'--sslPEMKeyFile', 'jstests/libs/server.pem',
'--sslCAFile', 'jstests/libs/ca.pem']
if "true" == check_output(run_mongo_cmd + ['--eval', 'print(db.runCommand("buildInfo").debug)']).strip():
_debug = True
if self.auth:
self.setup_admin_user(self.port)
if self.slave:
local = Connection(port=self.port, slave_okay=True).local
synced = False
while not synced:
synced = True
for source in local.sources.find(fields=["syncedTo"]):
synced = synced and "syncedTo" in source and source["syncedTo"]
print jsonstr
jsonpath = os.path.abspath(os.path.join(outfolder, 'world.txt'))
file = open(jsonpath, "w")
file.write(jsonstr)
file.close()
print 'DONE...'
if __name__ == "__main__":
if len(sys.argv) <> 3:
print "Usage: level-gen.py [input-svg-file] [output-folder]"
exit(1)
inputFn = os.path.abspath(sys.argv[1])
outputFolder = os.path.abspath(sys.argv[2])
utils.ensureDir(outputFolder)
print 'Level generator:'
print '----------------'
print 'Input Svg File: ' + inputFn
print 'Output Folder: ' + outputFolder
begin(inputFn, outputFolder)
def start(self):
global mongod_port
global mongod
if self.proc:
print >> sys.stderr, "probable bug: self.proc already set in start()"
return
self.ensure_test_dirs()
dir_name = smoke_db_prefix + "/data/db/sconsTests/"
self.port = int(mongod_port)
self.slave = False
if 'slave' in self.kwargs:
dir_name = smoke_db_prefix + '/data/db/sconsTestsSlave/'
srcport = mongod_port
self.port += 1
self.slave = True
if os.path.exists(dir_name):
pass
""" Cleanbb is the most irritating script ever created.
if 'slave' in self.kwargs:
argv = [utils.find_python(), "buildscripts/cleanbb.py", '--nokill', dir_name]
else:
argv = [utils.find_python(), "buildscripts/cleanbb.py", dir_name]
if quiet:
f = open(server_log_file, "a")
try:
call(argv, stdout=f)
finally:
f.close()
else:
call(argv)
"""
utils.ensureDir(dir_name)
argv = [mongod_executable, "--port", str(self.port), "--dbpath", dir_name]
if self.kwargs.get('small_oplog'):
argv += ["--master", "--oplogSize", "511"]
if self.kwargs.get('small_oplog_rs'):
argv += ["--replSet", "foo", "--oplogSize", "511"]
if self.slave:
argv += ['--slave', '--source', 'localhost:' + str(srcport)]
if self.kwargs.get('no_journal'):
argv += ['--nojournal']
if self.kwargs.get('no_preallocj'):
argv += ['--nopreallocj']
if self.kwargs.get('auth'):
argv += ['--auth']
self.auth = True
if len(server_log_file) > 0:
argv += ['--logpath', server_log_file]
if len(smoke_server_opts) > 0:
argv += [smoke_server_opts]
if not quiet:
print "running " + " ".join(argv)
self.proc = self._start(buildlogger(argv, is_global=True))
if not self.did_mongod_start(self.port):
raise Exception("Failed to start mongod")
if self.auth:
self.setup_admin_user(self.port)
if self.slave:
local = Connection(port=self.port, slave_okay=True).local
synced = False
while not synced:
synced = True
for source in local.sources.find(fields=["syncedTo"]):
synced = synced and "syncedTo" in source and source["syncedTo"]
def main():
# loading config file ...
cfgPath = sys.argv[1] if len(sys.argv) > 1 else './config.toml'
cfg = loadConfig(cfgPath)
try:
# ... and unpacking variables
dictget = lambda d, *k: [d[i] for i in k]
dataStats = cfg['data_stats']
modelParams = cfg['model_params']
trainCSV, testCSV = dictget(cfg['database'], 'train', 'test')
seqLength, stepSize = dictget(cfg['model_params'], 'seqLength',
'stepSize')
modelArch, modelDir, modelName = dictget(cfg['model_arch'],
'modelArch', 'modelDir',
'modelName')
optimizer, lossFunc, metricFuncs = dictget(cfg['training_params'],
'optimizer', 'lossFunc',
'metricFuncs')
lr, epochs, batchSize, patience, = dictget(cfg['training_params'],
'learningRate', 'epochs',
'batchSize', 'patience')
except KeyError as err:
print("\n\nERROR: not all parameters defined in config.toml : ", err)
print("Exiting ... \n\n")
sys.exit(1)
print("Loading training data ...")
xTrain, yTrain, stats = getData(trainCSV,
seqLength=seqLength,
stepSize=stepSize,
stats=dataStats)
print("Training Data Shape : ", xTrain.shape, "\n")
print("Loading testing data ...")
xTest, yTest, stats = getData(testCSV,
seqLength=seqLength,
stepSize=stepSize,
stats=dataStats)
print("Testing Data Shape : ", xTest.shape, "\n")
yTrain = np.expand_dims(
yTrain, -1) # adding extra axis as model expects 2 axis in the output
yTest = np.expand_dims(yTest, -1)
print("Compiling Model")
opt = getOptimizer(optimizer, lr)
model = makeModel(modelArch, modelParams, verbose=True)
model.compile(loss=lossFunc, optimizer=opt, metrics=metricFuncs)
# setting up directories
modelFolder = os.path.join(modelDir, modelName)
weightsFolder = os.path.join(modelFolder, "weights")
bestModelPath = os.path.join(weightsFolder, "best.hdf5")
ensureDir(bestModelPath)
saveConfig(cfgPath, modelFolder)
# callbacks
monitorMetric = 'val_loss'
check1 = ModelCheckpoint(os.path.join(weightsFolder,
modelName + "_{epoch:03d}.hdf5"),
monitor=monitorMetric,
mode='auto')
check2 = ModelCheckpoint(bestModelPath,
monitor=monitorMetric,
save_best_only=True,
mode='auto')
check3 = EarlyStopping(monitor=monitorMetric,
min_delta=0.01,
patience=patience,
verbose=0,
mode='auto')
check4 = CSVLogger(os.path.join(modelFolder,
modelName + '_trainingLog.csv'),
separator=',',
append=True)
check5 = ReduceLROnPlateau(monitor=monitorMetric,
factor=0.1,
patience=patience // 3,
verbose=1,
mode='auto',
min_delta=0.001,
cooldown=0,
min_lr=1e-10)
cb = [check2, check3, check4, check5]
if cfg['training_params']['saveAllWeights']:
cb.append(check1)
print("Starting Training ...")
model.fit(x=xTrain,
y=yTrain,
batch_size=batchSize,
epochs=epochs,
verbose=1,
callbacks=cb,
validation_data=(xTest, yTest),
shuffle=True)
def extractROIFromCenterAndSize(self,
roiCenterWorld,
roiSizeWorld,
magnification=None,
outputDir=None,
prefix=None,
oz=0,
flipX=False,
flipY=False,
removeTIFF=True,
makeNifti=True):
import nibabel as nib
import acsv
import utils
if magnification is None:
magnification = self.sourceLens
ijk2ras = self.getAffine(magnification)
ras2ijk = np.linalg.inv(ijk2ras)
roiCenterPixel = nib.affines.apply_affine(ras2ijk, roiCenterWorld)
sx, sy, sz, _ = np.diag(ijk2ras)
roiSizePixel = np.abs(roiSizeWorld / np.array((sx, sy, sz)))
topLeft = np.round(roiCenterPixel - roiSizePixel / 2).astype(int)
topRight = np.round(roiCenterPixel + roiSizePixel / 2).astype(int)
topRightX = topRight[0]
topLeftX, topLeftY, _ = topLeft
width, height, _ = np.round(roiSizePixel).astype(int)
if flipX:
topLeftX = self.flip(topRightX, 0, magnification)
topRightX = self.flip(topLeftX, 0, magnification)
if flipY:
topLeftY = self.flip(topLeftY, 1, magnification) - height
numTilesX = width / MAX_TILE_SIZE + 1
numTilesY = height / MAX_TILE_SIZE + 1
numTiles = numTilesX * numTilesY
tileWidth = width / numTilesX
tileHeight = height / numTilesY
if outputDir is None:
outputDir = os.path.dirname(self.filepath)
if prefix is None:
prefix = os.path.splitext(os.path.basename(self.filepath))[0]
for tileY in range(numTilesY):
for tileX in range(numTilesX):
tileTopLeftX = topLeftX + tileX * tileWidth
tileTopLeftY = topLeftY + tileY * tileHeight
tileTopRightX = tileTopLeftX + tileWidth - 1
if flipX:
tileColumn = numTilesX - tileX - 1
else:
tileColumn = tileX
if flipY:
tileRow = numTilesY - tileY - 1
else:
tileRow = tileY
if numTiles == 1:
outputPath = os.path.join(outputDir, prefix + '.tif')
else:
outputPath = os.path.join(
outputDir,
prefix + '_tile_%d_%d.tif' % (tileRow, tileColumn))
utils.ensureDir(outputPath)
roiPath = self.extractROI(magnification,
tileTopLeftX,
tileTopLeftY,
tileWidth,
tileHeight,
outputPath=outputPath)
if makeNifti:
import ImageUtils as iu
roiAffine = ijk2ras[:]
roiAffine[:3, :3] = np.abs(roiAffine[:3, :3])
roiAffine[0, 3] = self.flip(
tileTopRightX, 0, magnification) * roiAffine[0, 0]
roiAffine[1, 3] = tileTopLeftY * roiAffine[1, 1]
if flipX:
roiAffine[0, 0] *= -1
roiAffine[0, 3] = abs(
roiAffine[0, 0]) * (tileTopLeftX + tileWidth)
if flipY:
roiAffine[1, 1] *= -1
roiAffine[1, 3] = abs(roiAffine[1, 1]) * self.flip(
tileTopLeftY, 1, magnification)
roiAffine[2, 3] = oz
iu.histologyImageToNiftiRGB(roiPath, affine=roiAffine)
if removeTIFF:
os.remove(roiPath)