def load_test_data_ihm():
"""
Function to initialize MIMIC-III benchmark IHM test-data
Returns:
data (numpy.ndarray): discretized and normalized IHM data
labels (list[int]): true values of IHM task
names (list[str]): episode reference for true IHM value
discretizer_header (list[str]): header information on each variable
"""
discretizer = Discretizer(timestep=float(timestep),
store_masks=True,
impute_strategy='previous',
start_time='zero')
# Build readers, discretizers, normalizers
train_reader = InHospitalMortalityReader(
dataset_dir=os.path.join(data_dir, 'train'),
listfile=os.path.join(data_dir, 'train_listfile.csv'),
period_length=48.0)
discretizer_header = discretizer.transform(
train_reader.read_example(0)["X"])[1].split(',')
cont_channels = [
i for (i, x) in enumerate(discretizer_header) if x.find("->") == -1
]
normalizer = Normalizer(fields=cont_channels)
normalizer.load_params(normalizer_state)
test_reader = InHospitalMortalityReader(
dataset_dir=os.path.join(data_dir, 'test'),
listfile=os.path.join(data_dir, 'test_listfile.csv'),
period_length=48.0)
ret = utils.load_data(test_reader,
discretizer,
normalizer,
small_part,
return_names=True)
data = ret["data"][0]
labels = ret["data"][1]
names = ret["names"]
return data, labels, names, discretizer_header
listfile=os.path.join(args.data, 'train_listfile.csv'))
val_reader = InHospitalMortalityReader(
dataset_dir=os.path.join(args.data, 'train'),
listfile=os.path.join(args.data, 'val_listfile.csv'))
test_reader = InHospitalMortalityReader(
dataset_dir=os.path.join(args.data, 'test'),
listfile=os.path.join(args.data, 'test_listfile.csv'))
discretizer = Discretizer(timestep=args.timestep,
store_masks=True,
impute_strategy='previous',
start_time='zero')
discretizer_header = discretizer.transform(
train_reader.read_example(0)["X"])[1].split(',')
cont_channels = [
i for (i, x) in enumerate(discretizer_header) if x.find("->") == -1
]
normalizer = Normalizer(
fields=cont_channels) # choose here which columns to standardize
normalizer_state = 'ihm_ts{}.input_str:previous.start_time:zero.normalizer'.format(
args.timestep)
normalizer_state = os.path.join(os.path.dirname(__file__), normalizer_state)
normalizer.load_params(normalizer_state)
# Read data
train_raw = preprocessing.load_data(train_reader, discretizer, normalizer)
val_raw = preprocessing.load_data(val_reader, discretizer, normalizer)