def train(config, model, train_iter, dev_iter):
start_time = time.time()
model.train()
param_optimizer = list(model.named_parameters())
no_decay = ['bias', 'LayerNorm.bias', 'LayerNorm.weight']
optimizer_grouped_parameters = [
{'params': [p for n, p in param_optimizer if not any(nd in n for nd in no_decay)], 'weight_decay': 0.01},
{'params': [p for n, p in param_optimizer if any(nd in n for nd in no_decay)], 'weight_decay': 0.0}]
# optimizer = torch.optim.Adam(model.parameters(), lr=config.learning_rate)
optimizer = BertAdam(optimizer_grouped_parameters,
lr=config.learning_rate,
warmup=0.05,
t_total=len(train_iter) * config.num_epochs)
total_batch = 0 # 记录进行到多少batch
# dev_best_loss = float('inf')
dev_best_acc = -float('inf')
last_improve = 0 # 记录上次验证集loss下降的batch数
flag = False # 记录是否很久没有效果提升
model.train()
for epoch in range(config.num_epochs):
print('Epoch [{}/{}]'.format(epoch + 1, config.num_epochs))
for i, batch in enumerate(train_iter):
trains = (batch[0], batch[1], batch[2])
labels = torch.squeeze(batch[3], dim=1)
outputs = model(trains)
model.zero_grad()
class_weight = torch.FloatTensor([1, 1, 1, 1, 1, 1, 1, 0.4, 2, 1]).cuda()
loss = F.cross_entropy(outputs, labels, weight=class_weight)
# loss = F.cross_entropy(outputs, labels)
loss.backward()
optimizer.step()
if total_batch % 100 == 0:
# 每多少轮输出在训练集和验证集上的效果
true = labels.data.cpu()
predic = torch.max(outputs.data, 1)[1].cpu()
train_acc = metrics.accuracy_score(true, predic)
dev_acc, dev_loss, ouputs_all = evaluate(config, model, dev_iter)
if dev_acc > dev_best_acc:
# dev_best_loss = dev_loss
dev_best_acc = dev_acc
torch.save(model.state_dict(), config.save_path)
improve = '*'
last_improve = total_batch
else:
improve = ''
time_dif = get_time_dif(start_time)
msg = 'Iter: {0:>6}, Train Loss: {1:>5.2}, Train Acc: {2:>6.2%}, Val Loss: {3:>5.2}, Val Acc: {4:>6.2%}, Time: {5} {6}'
print(msg.format(total_batch, loss.item(), train_acc, dev_loss, dev_acc, time_dif, improve))
model.train()
total_batch += 1
if total_batch - last_improve > config.require_improvement:
# 验证集loss超过1000batch没下降，结束训练
print("No optimization for a long time, auto-stopping...")
flag = True
break
if flag:
break
def label_test(config, model, test_iter):
# test
model.load_state_dict(torch.load(config.save_path))
model.eval()
start_time = time.time()
test_acc, test_loss, test_report, test_confusion, _ = evaluate(config,
model,
test_iter,
test=True)
msg = 'Test Loss: {0:>5.2}, Test Acc: {1:>6.2%}'
print(msg.format(test_loss, test_acc))
print("Precision, Recall and F1-Score...")
print(test_report)
print("Confusion Matrix...")
print(test_confusion)
time_dif = get_time_dif(start_time)
print("Time usage:", time_dif)
seed = config.seed
np.random.seed(seed)
torch.manual_seed(seed)
torch.cuda.manual_seed_all(seed)
torch.backends.cudnn.deterministic = True # 保证每次结果一样
config.test_path = dataset + '/unlabeled_data.csv'
start_time = time.time()
data_df = load_data(config.test_path, config, with_label=False)
print('Reading testing data...')
test_data = Mydataset(config=config, data=data_df, with_labels=False)
test_iter = DataLoader(dataset=test_data,
batch_size=config.batch_size,
shuffle=False)
time_dif = get_time_dif(start_time)
print("Time usage:", time_dif)
model = bert_RNN(config).to(config.device)
predict_all = test(config, model, test_iter)
# ---------------------生成文件--------------------------
df_test = pd.read_csv(config.submit_example_path, encoding='utf-8')
id2label, label2id = json.load(open(config.id2label_path))
id2label = {int(i): j for i, j in id2label.items()} # 转为int型(原本是字符串形式)
class_labels = []
rank_labels = []
for i in predict_all:
label = str(id2label[i])
class_labels.append(label)
if label in ['财经', '时政']: