def execute_hmm(params):
config_prefix, config_suffix, data_file_base, num_support, crossval_num = params.split(
"___")
config_file = config_prefix + data_file_base + "_%s" % crossval_num + config_suffix
temp_dir = "temp_%s" % (crossval_num)
time.sleep(1 * int(crossval_num))
utils.remove_and_make_dir(temp_dir)
os.chdir(temp_dir)
HMM_command = ["./../HMM_EM", "Train", "../" + config_file
] # need to concatenate since we are running binary
results = subprocess.check_output(HMM_command)
test_data = None
for lead in range(1, 14):
try:
roc = run_inference_hmm.run_inference(data_file_base,
num_support,
"test",
lead,
plot_roc=False,
crossval=True,
crossval_num=crossval_num)
test_data = utils.add_to_data(test_data, [lead, roc])
except:
pass
os.chdir("..")
return np.atleast_2d(test_data)
def execute_hmm(params):
config_prefix, config_suffix, data_file_base, num_support, crossval_num = params.split("___")
config_file = config_prefix + data_file_base + "_%s" % crossval_num + config_suffix
temp_dir = "temp_%s" % (crossval_num)
time.sleep(1 * int(crossval_num))
utils.remove_and_make_dir(temp_dir)
os.chdir(temp_dir)
HMM_command = ["./../HMM_EM", "Train", "../" + config_file] # need to concatenate since we are running binary
results = subprocess.check_output(HMM_command)
test_data = None
for lead in range(1, 14):
try:
roc = run_inference_hmm.run_inference(
data_file_base, num_support, "test", lead, plot_roc=False, crossval=True, crossval_num=crossval_num
)
test_data = utils.add_to_data(test_data, [lead, roc])
except:
pass
os.chdir("..")
return np.atleast_2d(test_data)
def execute_hmm(params):
data_file_base, num_support, lead = params.split("___")
return run_inference_hmm.run_inference(data_file_base, int(num_support), "train", int(lead), plot_roc=False)
