def call(self, context):
args = context.args
input_path = args.inputData1
dicoms_path = os.path.join(input_path, "dicoms")
data = json.load(os.path.join(input_path, "data.json"))
try:
case, spacing, instances = read_dicom.load_dicom2(dicoms_path)
def call(self, context):
args = context.args
input_path = args.inputData1
input_npys_path = os.path.join(input_path, "npys")
data = json.load(os.path.join(input_path, "data.json"))
npys = [
os.path.join(input_npys_path, "case.npy"),
os.path.join(input_npys_path, "spacing.npy"),
os.path.join(input_npys_path, "prep_mask.npy"),
]
case, spacing, prep_mask = asyncio.map(npy.load,
npys,
thread=True,
pbar=True,
workers=len(npys))
try:
prep_data, extendbox, imgs, coord, nzhw = cpu_preprocess_2(
case, spacing, prep_mask, self.split_comber)
except FunctionTimedOut:
raise Exception("Preprocess timeout")
output_path = args.outputData1
output_npys_path = path.mkdirs(os.path.join(output_path, "npys"))
json.dump(data, os.path.join(output_path, "data.json"))
npys = [
(case, os.path.join(output_npys_path, "case.npy")),
(spacing, os.path.join(output_npys_path, "spacing.npy")),
(prep_data, os.path.join(output_npys_path, "prep_data.npy")),
(prep_mask, os.path.join(output_npys_path, "prep_mask.npy")),
(extendbox, os.path.join(output_npys_path, "extendbox.npy")),
(imgs.numpy(), os.path.join(output_npys_path, "imgs.npy")),
(coord.numpy(), os.path.join(output_npys_path, "coord.npy")),
(nzhw, os.path.join(output_npys_path, "nzhw.npy")),
]
asyncio.starmap(npy.dump,
npys,
thread=True,
pbar=True,
workers=len(npys))
return output_path
def get_balloon_dicts(img_dir, json_file):
"""
Parsing via json
"""
imgs_anns = json.load(json_file)
dataset_dicts = []
imagefile = [i.split(".jpg")[0] + ".jpg" for i in imgs_anns["metadata"].keys()]
for idx, v in enumerate(set(imagefile)):
record = {}
indices = [i for i, x in enumerate(imagefile) if x == v]
filename = os.path.join(img_dir, v)
height, width = image.read(filename).shape[:2]
record["file_name"] = filename
record["image_id"] = idx
record["height"] = height
record["width"] = width
objs = []
for index in indices:
data = list(imgs_anns["metadata"].values())[index]
xy = data["xy"][1:]
px = xy[::2]
py = xy[1::2]
poly = [(x + 0.5, y + 0.5) for x, y in zip(px, py)]
poly = list(itertools.chain.from_iterable(poly))
obj = {
"bbox": [np.min(px), np.min(py), np.max(px), np.max(py)],
"bbox_mode": BoxMode.XYXY_ABS,
"segmentation": [poly],
"category_id": 0,
"iscrowd": 0,
}
objs.append(obj)
record["annotations"] = objs
dataset_dicts.append(record)
return dataset_dicts
def call(self, context):
args = context.args
input_path = args.inputData1
npys_path = os.path.join(input_path, "npys")
data = json.load(os.path.join(input_path, "data.json"))
npys = [
os.path.join(npys_path, "case.npy"),
os.path.join(npys_path, "spacing.npy"),
os.path.join(npys_path, "prep_data.npy"),
os.path.join(npys_path, "prep_mask.npy"),
os.path.join(npys_path, "extendbox.npy"),
os.path.join(npys_path, "imgs.npy"),
os.path.join(npys_path, "coord.npy"),
os.path.join(npys_path, "nzhw.npy"),
]
case, spacing, prep_data, prep_mask, extendbox, imgs, coord, nzhw = asyncio.map(
npy.load, npys, thread=True, pbar=True, workers=len(npys))
imgs = torch.from_numpy(imgs)
coord = torch.from_numpy(coord)
try:
nodule_df, pbb = self.lung_dete.prediction(imgs,
coord,
nzhw,
spacing,
extendbox,
batch=self.args.batch)
preb = self.lung_isnc.nodule_cls(nodule_df, case, spacing)
preb = self.lung_lobe(preb, case, prep_mask, spacing)
data["nodule_preb"] = preb.to_dict()
logger.info("Sending predicted images...")
images = np.array([img for img in pickWithPbb(prep_data, pbb)])
return data, convert.flatAsImage(images)
except FunctionTimedOut:
raise Exception("Predict timeout")
def call(self, context):
args = context.args
input_path = args.inputData1
input_npys_path = os.path.join(input_path, "npys")
data = json.load(os.path.join(input_path, "data.json"))
npys = [
os.path.join(input_npys_path, "case.npy"),
os.path.join(input_npys_path, "spacing.npy"),
]
case, spacing = asyncio.map(npy.load,
npys,
thread=True,
pbar=True,
workers=len(npys))
try:
prep_mask = self.lung_segm.cut(case, 30)
except FunctionTimedOut:
raise Exception("Preprocess timeout")
output_path = args.outputData1
output_npys_path = path.mkdirs(os.path.join(output_path, "npys"))
json.dump(data, os.path.join(output_path, "data.json"))
npys = [
(case, os.path.join(output_npys_path, "case.npy")),
(spacing, os.path.join(output_npys_path, "spacing.npy")),
(prep_mask, os.path.join(output_npys_path, "prep_mask.npy")),
]
asyncio.starmap(npy.dump,
npys,
thread=True,
pbar=True,
workers=len(npys))
return output_path
def SPTrainer(context):
args = context.args
model = args.inputModel
device = "cuda" if args.__gpu else "cpu"
logger.info(
"**********Use {} Device for Training**********".format(device))
trainImg = args.inputTrainImage
trainJson = os.path.join(args.inputTrainJson, "project.json")
if args.inputTestImage is None or args.inputTestJson is None:
testImg = trainImg
else:
testImg = args.inputTestImage
if args.inputTestImage is None or args.inputTestJson is None:
testJson = trainJson
else:
testJson = os.path.join(args.inputTestJson, "project.json")
jsonData = json.load(trainJson)
classes = list(jsonData["attribute"]["1"]["options"].values())
model.dataset_register(trainImg, testImg, trainJson, testJson, classes)
model.device(device)
params = {
"IMS_PER_BATCH": args.ImgsPerBatch,
"BASE_LR": args.BaseLR,
"MAX_ITER": args.MaxIter,
"BATCH_SIZE_PER_IMAGE": args.BatchSizePerImg,
"NUM_CLASSES": len(classes),
"NUM_WORKERS": args.NumWorkers,
}
model.set_params(**params)
trainer = model.train()
model.evaluate(trainer, args.outputData)
return model, args.outputData