'threshold': None, # rs.pick_one(None, rs.uniform(0.0, 3.0)())(),
# classifier training /evaluation
'n_folds': 4, # since we can do this in parallel, it's best to have a multiple of the number of cores
}
TARGET_PATTERN = "/mnt/storage/usr/sedielem/whales/results/results-gen2-%s.pkl"
DATA_PATH = "/mnt/storage/usr/sedielem/whales/X_train.npy"
LABEL_PATH = "/mnt/storage/usr/sedielem/whales/Y_train.npy"
# TARGET_PATTERN = "results/results-gen2-%s.pkl"
# DATA_PATH = "X_train.npy"
# LABEL_PATH = "Y_train.npy"
expid = rs.generate_expid()
print "EXPERIMENT: %s" % expid
print
print settings
print
start_time = time.time()
def tock():
elapsed = time.time() - start_time
print " running for %.2f s" % elapsed
# load data
print "Load data"
X = np.load(DATA_PATH)
'threshold': None,
# classifier training /evaluation
'n_folds':
4, # since we can do this in parallel, it's best to have a multiple of the number of cores
}
# TARGET_PATTERN = "/mnt/storage/usr/sedielem/whales/results/results-%s.pkl"
# DATA_PATH = "/mnt/storage/usr/sedielem/whales/X_train.npy"
# LABEL_PATH = "/mnt/storage/usr/sedielem/whales/Y_train.npy"
TARGET_PATTERN = "results/results-%s.pkl"
DATA_PATH = "X_train.npy"
LABEL_PATH = "Y_train.npy"
expid = rs.generate_expid()
print("EXPERIMENT: %s" % expid)
print()
print(settings)
print()
start_time = time.time()
def tock():
elapsed = time.time() - start_time
print(" running for %.2f s" % elapsed)
# load data
print("Load data")