def forward(self, seq_dignosis_codes, seq_time_step, batch_labels, options,
maxlen):
seq_time_step = np.array(list(units.pad_time(seq_time_step, options)))
lengths = torch.from_numpy(
np.array([len(seq) for seq in seq_dignosis_codes])).cuda()
diagnosis_codes, labels, mask, mask_final, mask_code = units.pad_matrix_new(
seq_dignosis_codes, batch_labels, options)
if options['use_gpu']:
diagnosis_codes = torch.LongTensor(diagnosis_codes).cuda()
mask_mult = torch.BoolTensor(1 - mask).unsqueeze(2).cuda()
mask_final = torch.Tensor(mask_final).unsqueeze(2).cuda()
mask_code = torch.Tensor(mask_code).unsqueeze(3).cuda()
else:
diagnosis_codes = torch.LongTensor(diagnosis_codes)
mask_mult = torch.BoolTensor(1 - mask).unsqueeze(2)
mask_final = torch.Tensor(mask_final).unsqueeze(2)
mask_code = torch.Tensor(mask_code).unsqueeze(3)
features = self.feature_encoder(diagnosis_codes, mask_mult, mask_code,
seq_time_step, lengths)
final_statues = features * mask_final
final_statues = final_statues.sum(1, keepdim=True)
quiryes = self.relu(self.quiry_layer(final_statues))
self_weight = self.get_self_attention(features, quiryes, mask_mult)
total_weight = self_weight
total_weight = total_weight / (
torch.sum(total_weight, 1, keepdim=True) + 1e-5)
weighted_features = features * total_weight
averaged_features = torch.sum(weighted_features, 1)
averaged_features = self.dropout(averaged_features)
predictions = self.classify_layer(averaged_features)
labels = torch.LongTensor(labels)
if options['use_gpu']:
labels = labels.cuda()
return predictions, labels, self_weight
def forward(self, seq_dignosis_codes, seq_time_step, batch_labels, options,
maxlen):
seq_time_step = np.array(list(units.pad_time(seq_time_step, options)))
lengths = torch.from_numpy(
np.array([len(seq) for seq in seq_dignosis_codes])).cuda()
diagnosis_codes, labels, mask, mask_final, mask_code = units.pad_matrix_new(
seq_dignosis_codes, batch_labels, options)
if options['use_gpu']:
diagnosis_codes = torch.LongTensor(diagnosis_codes).cuda()
mask_mult = torch.BoolTensor(1 - mask).unsqueeze(2).cuda()
mask_final = torch.Tensor(mask_final).unsqueeze(2).cuda()
mask_code = torch.Tensor(mask_code).unsqueeze(3).cuda()
else:
diagnosis_codes = torch.LongTensor(diagnosis_codes)
mask_mult = torch.BoolTensor(1 - mask).unsqueeze(2)
mask_final = torch.Tensor(mask_final).unsqueeze(2)
mask_code = torch.Tensor(mask_code).unsqueeze(3)
features = self.feature_encoder(diagnosis_codes, mask_mult, mask_code,
seq_time_step, lengths)
final_statues = features * mask_final
final_statues = final_statues.sum(1)
predictions = self.classify_layer(final_statues)
labels = torch.LongTensor(labels)
if options['use_gpu']:
labels = labels.cuda()
return predictions, labels, None
def load_hita_input(input_jsonl_path, max_seq_len):
with open("./data/hfdata/hf_code2idx_new.pickle", "rb") as fin:
code2id = pickle.load(fin)
n_diagnosis_codes = len(code2id)
diagnosis_codes = []
labels = []
qids = []
time_step = []
with open(input_jsonl_path, "r", encoding="utf-8") as fin:
for line in fin:
input_json = json.loads(line)
qids.append(input_json['id'])
record_icd = input_json["medical_records"]["record_icd"]
time_dis = input_json["medical_records"]["time_distance"]
label = input_json["heart_diseases"]["hf_label"]
for i in range(len(record_icd)):
for j in range(len(record_icd[i])):
record_icd[i][j] = code2id[record_icd[i][j]]
diagnosis_codes.append(record_icd)
time_step.append(time_dis)
labels.append(label)
# diagnosis_codes = np.array(diagnosis_codes)
# time_step = np.array(time_step)
diagnosis_codes, time_step = units.adjust_input(diagnosis_codes, time_step,
max_seq_len,
n_diagnosis_codes)
lengths = np.array([max_seq_len + 1 for seq in diagnosis_codes])
seq_time_step = np.array(list(units.pad_time(time_step, max_seq_len + 1)))
lengths = torch.from_numpy(lengths)
diagnosis_codes, mask, mask_final, mask_code = units.pad_matrix_new(
diagnosis_codes, n_diagnosis_codes, max_seq_len + 1)
diagnosis_codes = torch.LongTensor(diagnosis_codes)
mask_mult = torch.ByteTensor(1 - mask).unsqueeze(2)
mask_final = torch.Tensor(mask_final).unsqueeze(2)
mask_code = torch.Tensor(mask_code).unsqueeze(3)
seq_time_step = torch.Tensor(seq_time_step).unsqueeze(2) / 180
labels = torch.tensor(labels, dtype=torch.long)
diagnosis_codes, seq_time_step, mask_mult, mask_final, mask_code, lengths = \
diagnosis_codes.unsqueeze(1), seq_time_step.unsqueeze(1), mask_mult.unsqueeze(1), \
mask_final.unsqueeze(1), mask_code.unsqueeze(1), lengths.unsqueeze(1)
return qids, labels, diagnosis_codes, seq_time_step, mask_mult, mask_final, mask_code, lengths