# del train_df
gc.collect()
print('preparing indexes and group')
train_indexes = list(group.index)[-TRAIN_SAMPLES:]
valid_indexes = list(group.index)[:-TRAIN_SAMPLES]
train_group = group[group.index.isin(train_indexes)]
valid_group = group[group.index.isin(valid_indexes)]
del group, train_indexes, valid_indexes
print(len(train_group), len(valid_group))
# print('preparing questions concepts')
# q_concepts = load_questions_part_and_tags('./data/questions.csv', params.n_question)
print('preparing training dataloader')
train_dataset = DS.AKTDataset(train_group, n_skill, max_seq=params.max_seq)
train_dataloader = DataLoader(train_dataset,
batch_size=params.batch_size,
shuffle=True,
num_workers=12)
del train_group
print('preparing validation dataloader')
valid_dataset = DS.AKTDataset(valid_group, n_skill, max_seq=params.max_seq)
valid_dataloader = DataLoader(valid_dataset,
batch_size=params.batch_size,
shuffle=False,
num_workers=12)
del valid_group
###############################
