def prepare_folds(hdf5, folds, pheno, derivatives, experiment):
# 创建实验数据
exps = hdf5.require_group("experiments")
ids = pheno["FILE_ID"]
for derivative in derivatives:
# 为每个脑图谱创建一个实验数据
exp = exps.require_group(
format_config(experiment, {
"derivative": derivative,
}))
exp.attrs["derivative"] = derivative
# 按照标签比例把数据分为n份，并把数据进行打散
skf = StratifiedKFold(n_splits=folds, shuffle=True)
for i, (train_index,
test_index) in enumerate(skf.split(ids, pheno["STRAT"])):
train_index, valid_index = train_test_split(train_index,
test_size=0.33)
# 创建每个分组的实验数据
fold = exp.require_group(str(i))
fold['train'] = [ind.encode('utf8') for ind in ids[train_index]]
fold['valid'] = [indv.encode('utf8') for indv in ids[valid_index]]
fold["test"] = [indt.encode('utf8') for indt in ids[test_index]]
def load_patient(subj, tmpl):
# 拼接参数获取数据地址，并读取数据
df = pd.read_csv(format_config(tmpl, {
"subject": subj,
}),
sep="\t",
header=0)
df = df.apply(lambda x: pd.to_numeric(x, errors='coerce'))
# 获取ROI区域编号
ROIs = [
"#" + str(y) for y in sorted([int(x[1:]) for x in df.keys().tolist()])
]
if arguments["--lstm"]:
functional = np.nan_to_num(df[ROIs].to_numpy()).tolist()
else:
# 使用0替代无效元素，一共200行，表示200个感兴趣区域，每行一共有196个元素，表示每个感兴趣区域有196个值
functional = np.nan_to_num(df[ROIs].to_numpy().T).tolist()
# axis=1表示沿着x轴数据标准化
functional = preprocessing.scale(functional, axis=1)
if arguments["--lstm"]:
functional = np.array(functional)
else:
# 计算并获得每两个ROI之间的连接性
functional = compute_connectivity(functional)
functional = functional.astype(np.float32)
# 返回某个病人的ROI的连接性
return subj, functional.tolist()
load_patients_to_file(hdf5, pheno, derivatives)
# 构建所有的交叉验证的实验数据
if arguments["--whole"]:
print ("Preparing whole dataset")
prepare_folds(hdf5, folds, pheno, derivatives, experiment="{derivative}_whole")
# 构建男性的交叉验证的实验数据
if arguments["--male"]:
print ("Preparing male dataset")
pheno_male = pheno[pheno["SEX"] == "M"]
prepare_folds(hdf5, folds, pheno_male, derivatives, experiment="{derivative}_male")
# 构建有阈值的交叉验证的实验数据
if arguments["--threshold"]:
print ("Preparing thresholded dataset")
pheno_thresh = pheno[pheno["MEAN_FD"] <= 0.2]
prepare_folds(hdf5, folds, pheno_thresh, derivatives, experiment="{derivative}_threshold")
# 构建每个实验室的交叉验证的实验数据
if arguments["--leave-site-out"]:
print ("Preparing leave-site-out dataset")
for site in pheno["SITE_ID"].unique():
pheno_without_site = pheno[pheno["SITE_ID"] != site]
prepare_folds(hdf5, folds, pheno_without_site, derivatives, experiment=format_config(
"{derivative}_leavesiteout-{site}",
{
"site": site,
})
)
pheno_male = pheno[pheno["SEX"] == "M"]
prepare_folds(hdf5,
folds,
pheno_male,
derivatives,
experiment="{derivative}_male")
# 构建有阈值的交叉验证的实验数据
if arguments["--threshold"]:
print("Preparing thresholded dataset")
pheno_thresh = pheno[pheno["MEAN_FD"] <= 0.2]
prepare_folds(hdf5,
folds,
pheno_thresh,
derivatives,
experiment="{derivative}_threshold")
# 构建每个实验室的交叉验证的实验数据
if arguments["--leave-site-out"]:
print("Preparing leave-site-out dataset")
for site in pheno["SITE_ID"].unique():
pheno_without_site = pheno[pheno["SITE_ID"] != site]
prepare_folds(hdf5,
folds,
pheno_without_site,
derivatives,
experiment=format_config(
"{derivative}_leavesiteout-{site}", {
"site": site,
}))
# 脑图谱的选择
valid_derivatives = ["cc200", "aal", "ez", "ho", "tt", "dosenbach160"]
derivatives = [
derivative for derivative in arguments["<derivative>"]
if derivative in valid_derivatives
]
# 标记实现数据
experiments = []
for derivative in derivatives:
config = {"derivative": derivative}
if arguments["--whole"]:
experiments += [
PrepareUtils.format_config("{derivative}_whole", config)
],
if arguments["--male"]:
experiments += [
PrepareUtils.format_config("{derivative}_male", config)
]
if arguments["--threshold"]:
experiments += [
PrepareUtils.format_config("{derivative}_threshold", config)
]
if arguments["--leave-site-out"]:
for site in pheno["SITE_ID"].unique():
site_config = {"site": site}