# Random splitting of data to train our model
X_train, X_test, y_train, y_test = train_test_split(new_ukbb,
y,
test_size=0.5,
random_state=0)
X_train = X_train.dropna(subset=['21003-2.0'])
X_test = X_test.dropna(subset=['21003-2.0'])
merged_data = pd.read_csv(path_to_merge_brain, usecols=columns)
dmriDict = collections.OrderedDict(
chain(brain_dmri_fa.items(), brain_dmri_icvf.items(),
brain_dmri_isovf.items(), brain_dmri_l1.items(),
brain_dmri_l2.items(), brain_dmri_l3.items(), brain_dmri_md.items(),
brain_dmri_mo.items(), brain_dmri_od.items()))
dmriDict.update({'eid': 'eid'})
dmri = pd.DataFrame(merged_data, columns=dmriDict.keys())
dmri = dmri.dropna()
def load_combine_data(X_split, merged_data, dmri):
data_frame = []
connectomes = []
eids = []
for e_id in X_split.eid:
this_eid_data = merged_data[merged_data['eid'] == e_id]
this_path = os.path.join(path_to_matrices,
str(e_id) + '_20227_2_0.txt')
this_dmri_data = dmri[dmri['eid'] == e_id]
# Random splitting of data to train our model
X_train, X_test, y_train, y_test = train_test_split(
new_ukbb, y, test_size=0.5, random_state=0)
X_train = X_train[['eid', '20127-0.0']].dropna()
X_test = X_test[['eid', '20127-0.0']].dropna()
merged_data = pd.read_csv(path_to_merge_brain, usecols=columns)
dmriDict = collections.OrderedDict(chain(brain_dmri_fa.items(),
brain_dmri_icvf.items(),
brain_dmri_isovf.items(),
brain_dmri_l1.items(),
brain_dmri_l2.items(),
brain_dmri_l3.items(),
brain_dmri_md.items(),
brain_dmri_mo.items(),
brain_dmri_od.items()))
dmriDict.update({'eid': 'eid'})
dmri = pd.DataFrame(merged_data, columns=dmriDict.keys())
dmri = dmri.dropna()
def load_combine_data(X_split, merged_data, dmri):
data_frame = []
eids = []
for e_id in X_split.eid:
this_eid_data = merged_data[merged_data['eid'] == e_id]
this_dmri_data = dmri[dmri['eid'] == e_id]
if not e_id == 3551399:
if not len(this_dmri_data) == 0: