print("StdCall Extractor started!")
# Training set or test set?
if sys.argv[1] == "train":
main_path = os.path.join(root_path, "train")
else:
main_path = os.path.join(root_path, "test")
# Generate paths
data_dir = os.path.join(main_path, os.path.join("data", "stdcall"))
data = [
os.path.join(data_dir, file) for file in os.listdir(data_dir)
if os.path.isfile(os.path.join(data_dir, file))
]
split_data = chunks(data, numWorkers)
# Directory name
output_dir = os.path.join(main_path, stdcall_feat_dir)
clearPath(output_dir)
# Parallel threading
jobs = []
for item in enumerate(split_data):
p = Process(target=worker, args=(item, output_dir))
jobs.append(p)
p.start()
for j in jobs:
j.join()
# Save features
pickle.dump(feat, open(os.path.join(output_dir, fn), "wb"))
# Update progressbar
pbar.update(idx + 1)
pbar.finish()
if __name__ == "__main__":
print("Feature extractor started!")
# Which dataset? Train or test?
if sys.argv[1] == "train":
main_path = os.path.join(root_path, "train")
split_data = chunks(train_asm, numWorkers)
else:
main_path = os.path.join(root_path, "test")
split_data = chunks(test_asm, numWorkers)
# Directory name
output_dir = os.path.join(main_path, feat_dir)
clearPath(output_dir)
# Multi threading
jobs = []
for item in enumerate(split_data):
p = Process(target=worker, args=(item, output_dir))
jobs.append(p)
p.start()
# Save data
data = calcGistFeatures(r.readBytes(item), kernels)
pickle.dump(data, open(os.path.join(output_dir, fn), "wb"))
# Update progressbar
pbar.update(idx + 1)
pbar.finish()
if __name__ == "__main__":
print("Byte image features extraction started!")
if sys.argv[1] == "train":
main_path = os.path.join(root_path, "train")
split_data = chunks(train_bytes, numWorkers)
else:
main_path = os.path.join(root_path, "test")
split_data = chunks(test_bytes, numWorkers)
output_dir = os.path.join(main_path, byte_image_feat_dir)
clearPath(output_dir)
jobs = []
for item in enumerate(split_data):
p = Process(target=worker, args=(item, output_dir))
jobs.append(p)
p.start()
for j in jobs: