import SimpleITK as sitk
#######################
# Parse the command line
from neon.util.argparser import NeonArgparser
parser = NeonArgparser(__doc__)
# We can pass the input directory and output file name from the command line
parser.add_argument('-out',
'--outFilename',
default='dicom_out.h5',
help='Name of the output HDF5 file')
parser.set_defaults(data_dir='/Volumes/data/tonyr/dicom/Lung CT/stage1')
parser.set_defaults(save_path='.')
args = parser.parse_args()
data_dir = args.data_dir
outFilename = args.save_path + '/' + args.outFilename
##############################
def verbosePrint(txt):
print(txt)
verbosePrint('DICOM to HDF5 converter started ... ')
output (str): output file path
predictions:
the model's predictions
"""
results_list = predictions.tolist()
with open(output, 'w', encoding='utf-8') as out_file:
writer = csv.writer(out_file, delimiter=',', quotechar='"')
for result in results_list:
writer.writerow([result])
print("Results of inference saved in {0}".format(output))
if __name__ == "__main__":
# parse the command line arguments
parser = NeonArgparser()
parser.set_defaults(epochs=200)
parser.add_argument('--data', help='prepared data CSV file path',
type=validate_existing_filepath)
parser.add_argument('--model', help='path to the trained model file',
type=validate_existing_filepath)
parser.add_argument('--print_stats', action='store_true', default=False,
help='print evaluation stats for the model predictions - if '
'your data has tagging')
parser.add_argument('--output', help='path to location for inference output file',
type=validate_parent_exists)
args = parser.parse_args()
data_path = absolute_path(args.data)
model_path = absolute_path(args.model)
print_stats = args.print_stats
output_path = absolute_path(args.output)
# generate backend
parser.set_defaults(
#constant arguments
rng_seed=2,
backend= "cpu",
progress_bar=True,
verbose=4,
evaluation_freq=2,
#data
epochs= 10,
batch_size=128,
data_dir="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/DATA/",
file_name="hurricanes.h5",
nclass=2,
data_dict=["1","0"],
norm_type=2, #1: global contrast norm, 2:standard norm, 3:l1/l2 norm, scikit learn
train_num_p=8000,
valid_num_p=1000,
test_num_p=1000,
train_num_n=8000,
valid_num_n=1000,
test_num_n=1000,
# results
#out_dir="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/hurricane/",
#save_path="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/hurricane/hurricane_classify_train.pkl",
#serialize= 2,
#logfile="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/hurricane/hurricane_classify_train."+c_time+".log",
#output_file="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/hurricane/hurricane_classify_train."+c_time+".h5",
)
parser.set_defaults(
#constant arguments
rng_seed=2,
backend= "cpu",
progress_bar=True,
verbose=4,
evaluation_freq=2,
#data
epochs= 2,
batch_size=100,
data_dir="./TEST_DATA",
file_name="hurricanes.h5",
nclass=2,
data_dict=["1","0"],
norm_type=2, #1: global contrast norm, 2:standard norm, 3:l1/l2 norm, scikit learn
train_num_p=1000,
valid_num_p=1000,
test_num_p=1000,
train_num_n=1000,
valid_num_n=1000,
test_num_n=1000,
# results
out_dir="./TEST_RESULTS",
save_path="./TEST_RESULTS/hurricane_classify_train.pkl",
serialize= 2,
logfile="./TEST_RESULTS/hurricane_classify_train."+c_time+".log",
output_file="./TEST_RESULTS/hurricane_classify_train."+c_time+".h5",
)
parser.set_defaults(
#constant arguments
rng_seed=2,
backend= "cpu",
dataype="f32",
progress_bar=True,
verbose=4,
#evaluation_freq=3,
#variable arguments
epochs= 10,
batch_size=100,
#data_dir="/global/project/projectdirs/nervana/yunjie/climatedata/new_landsea/",
#file_name="atmosphericriver_us+TMQ+land_Sep4.h5",
data_dir="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/DATA/",
file_name="atmospheric_river_us+eu+landsea_sep10.h5",
nclass=2, #number of event category to classify,
data_dict=["AR","Non_AR"],
norm_type=3, #1: global contrast norm, 2:standard norm, 3:l1/l2 norm, scikit learn
#TODO, make the "nclass" reading from input files, more general
train_num_p=5000, #positive training example
valid_num_p=800,
test_num_p=1000,
train_num_n=5000, #negative training example
valid_num_n=500,
test_num_n=1000,
#output files
#out_dir="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/atmosphericriver/",
#save_path="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/atmosphericriver/ar_classify_train.pkl",
#serialize= 2,
#logfile="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/atmosphericriver/ar_classify_train."+c_time+".log",
#output_file="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/atmosphericriver/ar_classify_train."+c_time+".h5",
)
for a in argss:
parser.add_argument(a)
parser.set_defaults(
#constant arguments
#rng_seed=2,
backend= "cpu",
dataype="f32",
progress_bar=True,
log_thresh=10,
#variable arguments
epochs= 15,
batch_size=100,
data_dir="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/DATA/",
file_name="fronts_all.h5",
nclass=2,
data_dict=["Front","NonFront"],
norm_type=2, #1: global contrast norm, 2:standard norm, 3:l1/l2 norm, scikit learn
#TODO, make the "nlcass" reading from input data, more general
#follow 80% 20% rule below
train_num_p=4000,
valid_num_p=600,
test_num_p=1000,
train_num_n=4000,
valid_num_n=600,
test_num_n=1000
)
args = parser.parse_args()
parser.set_defaults(
#constant arguments
rng_seed=2,
backend= "cpu",
progress_bar=True,
verbose=4,
evaluation_freq=2,
#data
epochs= 1,
batch_size=100, ####when testing, make the batch size equal to test data length (easier for later confusion matrix and feature sample)
data_dir="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/DATA/",
file_name="hurricanes.h5",
nclass=2,
data_dict=["1","0"],
norm_type=2, #1: global contrast norm, 2:standard norm, 3:l1/l2 norm, scikit learn
train_num_p=8000,
valid_num_p=1000,
test_num_p=1000,
train_num_n=8000,
valid_num_n=1000,
test_num_n=1000,
# results
#out_dir="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/",
#save_path="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/hurricane_classify_train_S.pkl",
#serialize= 2,
logfile="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/hurricane/hurricane_classify_test."+c_time+".log",
#output_file="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/hurricane_classify_S."+c_time+".h5",
model_file="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/hurricane/hurricane_classify_train.pkl",
)
parser.set_defaults(
#constant arguments
rng_seed=3,
backend= "cpu",
dataype="f32",
progress_bar=True,
verbose=4,
#evaluation_freq=1,
#variable arguments
epochs= 15,
batch_size=100,
data_dir="/global/project/projectdirs/nervana/yunjie/climatedata/old/",
file_name="fronts_all.h5",
nclass=2,
data_dict=["Front","NonFront"],
norm_type=2, #1: global contrast norm, 2:standard norm, 3:l1/l2 norm, scikit learn
#TODO, make the "nlcass" reading from input data, more general
#follow 80% 20% rule below
train_num_p=4000,
valid_num_p=600,
test_num_p=1000,
train_num_n=4000,
valid_num_n=600,
test_num_n=1000,
# results
#out_dir="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/fronts/",
#save_path="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/fronts/fronts_classify_train.pkl",
#serialize= 2,
#logfile="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/fronts/fronts_classify_train."+c_time+".log",
#output_file="/global/project/projectdirs/nervana/yunjie/climate_neon1.0run/conv/RESULTS/fronts/fronts_classify_train."+c_time+".h5",
)
