def evaluate(model, Loss, x_val, y_val):
"""测试集上准确率评估"""
batch_val = batch_iter(x_val, y_val, 64)
acc = 0
los = 0
model.eval()
with torch.no_grad():
for x_batch, y_batch in batch_val:
size = len(x_batch)
x = np.array(x_batch)
y = np.array(y_batch)
x = torch.LongTensor(x)
y = torch.Tensor(y)
# y = torch.LongTensor(y)
# x = Variable(x)
# y = Variable(y)
out = model(x)
loss = Loss(out, y)
# optimizer.zero_grad()
# loss.backward()
# optimizer.step()
loss_value = np.mean(loss.numpy())
accracy = np.mean((torch.argmax(out,
1) == torch.argmax(y, 1)).numpy())
acc += accracy * size
los += loss_value * size
model.train()
return los / len(x_val), acc / len(x_val)
def evaluate(sess,x_,y_):
data_len=len(x_)
batch_eval=batch_iter(x_,y_,128)
total_loss=0.0
total_acc=0.0
for x_batch,y_batch in batch_eval:
batch_len=len(x_batch)
feed_dict=feed_data(x_batch,y_batch,1.0)
loss,acc=sess.run([model.loss,model.acc],feed_dict=feed_dict)
total_loss+=loss*batch_len
total_acc+=acc*batch_len
return total_loss/data_len,total_acc/data_len
def evaluate(sess, x_, y_):
"""评估在某一数据上的准确率和损失"""
data_len = len(x_)
batch_eval = batch_iter(x_, y_, 128)
total_loss = 0.0
total_acc = 0.0
for x_batch, y_batch in batch_eval:
batch_len = len(x_batch)
feed_dict = feed_data(x_batch, y_batch, 1.0)
y_pred_class,loss, acc = sess.run([model.y_pred_cls,model.loss, model.acc], feed_dict=feed_dict)
total_loss += loss * batch_len
total_acc += acc * batch_len
return y_pred_class,total_loss / data_len, total_acc / data_len
def train():
x_train, y_train = process_file(train_dir, word_to_id, cat_to_id,
600) #获取训练数据每个字的id和对应标签的oe-hot形式
x_val, y_val = process_file(val_dir, word_to_id, cat_to_id, 600)
#使用LSTM或者CNN
model = TextRNN()
model.train()
# model = TextCNN()
#选择损失函数
Loss = nn.MultiLabelSoftMarginLoss()
# Loss = nn.BCELoss()
# Loss = nn.MSELoss()
optimizer = optim.Adam(model.parameters(), lr=0.001)
best_val_acc = 0
for epoch in range(100):
i = 0
print('epoch:{}'.format(epoch))
batch_train = batch_iter(x_train, y_train, 64)
for x_batch, y_batch in batch_train:
i += 1
# print(i)
x = np.array(x_batch)
y = np.array(y_batch)
x = torch.LongTensor(x)
y = torch.Tensor(y)
# y = torch.LongTensor(y)
# x = Variable(x)
# y = Variable(y)
out = model(x)
loss = Loss(out, y)
optimizer.zero_grad()
loss.backward()
optimizer.step()
# 对模型进行验证
if i % 90 == 0:
los, accracy = evaluate(model, Loss, x_val,
y_val) # 此处不需要优化器参数
print('loss:{},accracy:{}'.format(los, accracy))
if accracy > best_val_acc:
torch.save(model.state_dict(), 'model_params.pkl')
best_val_acc = accracy
def train():
print("Configuring TensorBoard and Saver...")
# 配置 Tensorboard，重新训练时，请将tensorboard文件夹删除，不然图会覆盖
tensorboard_dir = 'tensorboard/textrnn'
if not os.path.exists(tensorboard_dir):
os.makedirs(tensorboard_dir)
tf.summary.scalar("loss", model.loss)
tf.summary.scalar("accuracy", model.acc)
merged_summary = tf.summary.merge_all()
writer = tf.summary.FileWriter(tensorboard_dir)
# 配置 Saver
saver = tf.train.Saver()
if not os.path.exists(save_dir):
os.makedirs(save_dir)
print("Loading training and validation data...")
# 载入训练集与验证集
start_time = time.time()
x_train, y_train = process_file(train_dir, word_to_id, cat_to_id, config.seq_length)
x_val, y_val = process_file(val_dir, word_to_id, cat_to_id, config.seq_length)
time_dif = get_time_dif(start_time)
print("Time usage:", time_dif)
# 创建session
session = tf.Session()
session.run(tf.global_variables_initializer())
writer.add_graph(session.graph)
print('Training and evaluating...')
start_time = time.time()
total_batch = 0 # 总批次
best_acc_val = 0.0 # 最佳验证集准确率
last_improved = 0 # 记录上一次提升批次
require_improvement = 1000 # 如果超过1000轮未提升，提前结束训练
flag = False
for epoch in range(config.num_epochs):
print('Epoch:', epoch + 1)
batch_train = batch_iter(x_train, y_train, config.batch_size)
for x_batch, y_batch in batch_train:
feed_dict = feed_data(x_batch, y_batch, config.dropout_keep_prob)
if total_batch % config.save_per_batch == 0:
# 每多少轮次将训练结果写入tensorboard scalar
s = session.run(merged_summary, feed_dict=feed_dict)
writer.add_summary(s, total_batch)
if total_batch % config.print_per_batch == 0:
# 每多少轮次输出在训练集和验证集上的性能
feed_dict[model.keep_prob] = 1.0
loss_train, acc_train = session.run([model.loss, model.acc], feed_dict=feed_dict)
loss_val, acc_val = evaluate(session, x_val, y_val) # todo
if acc_val > best_acc_val:
# 保存最好结果
best_acc_val = acc_val
last_improved = total_batch
saver.save(sess=session, save_path=save_path)
improved_str = '*'
else:
improved_str = ''
time_dif = get_time_dif(start_time)
msg = 'Iter: {0:>6}, Train Loss: {1:>6.2}, Train Acc: {2:>7.2%},' \
+ ' Val Loss: {3:>6.2}, Val Acc: {4:>7.2%}, Time: {5} {6}'
print(msg.format(total_batch, loss_train, acc_train, loss_val, acc_val, time_dif, improved_str))
session.run(model.optim, feed_dict=feed_dict) # 运行优化
total_batch += 1
if total_batch - last_improved > require_improvement:
# 验证集正确率长期不提升，提前结束训练
print("No optimization for a long time, auto-stopping...")
flag = True
break # 跳出循环
if flag: # 同上
break
def train():
print("Configuring TensorBoard and Saver...")
tensorboard_dir = 'tensorboard/textcnn'
if not os.path.exists(tensorboard_dir):
os.makedirs(tensorboard_dir)
tf.summary.scalar("loss", model.loss)
tf.summary.scalar("accuracy", model.acc)
# 用到 tf.summary 中的方法保存日志数据，用于tensorboard可视化操作。
# 用 tf.summary.scalar 保存标量，一般用来保存loss，accuary，学习率等数据
merged_summary = tf.summary.merge_all()
writer = tf.summary.FileWriter(tensorboard_dir)
# 使用 tf.summaries.merge_all() 对所有的汇总操作进行合并
# 将数据写入本地磁盘: tf.summary.FileWriter
saver = tf.train.Saver()
if not os.path.exists(save_dir):
os.makedirs(save_dir)
print("Loading training and validation data...")
start_time = time.time()
x_train, y_train = process_file(train_dir, word_to_id, cat_to_id, config.seq_length)
x_val, y_val = process_file(val_dir, word_to_id, cat_to_id, config.seq_length)
time_dif = get_time_dif(start_time)
print("Time usage:", time_dif)
session = tf.Session()
session.run(tf.global_variables_initializer())
writer.add_graph(session.graph)
print('Training and evaluating...')
start_time = time.time()
total_batch = 0
best_acc_val = 0.0
last_improved = 0
require_improvement = 1000
# 如果超过1000轮未提升，提前结束训练，防止过拟合。
flag = False
for epoch in range(config.num_epochs):
print('Epoch:', epoch + 1)
batch_train = batch_iter(x_train, y_train, config.batch_size)
for x_batch, y_batch in batch_train:
feed_dict = feed_data(x_batch, y_batch, config.dropout_keep_prob)
if total_batch % config.save_per_batch == 0:
s = session.run(merged_summary, feed_dict=feed_dict)
writer.add_summary(s, total_batch)
if total_batch % config.print_per_batch == 0:
feed_dict[model.keep_prob] = 1.0
loss_train, acc_train = session.run([model.loss, model.acc], feed_dict=feed_dict)
loss_val, acc_val = evaluate(session, x_val, y_val) # todo
if acc_val > best_acc_val:
# 保存最好结果
best_acc_val = acc_val
last_improved = total_batch
saver.save(sess=session, save_path=save_path)
improved_str = '*'
else:
improved_str = ''
time_dif = get_time_dif(start_time)
msg = 'Iter: {0:>6}, Train Loss: {1:>6.2}, Train Acc: {2:>7.2%},' \
+ ' Val Loss: {3:>6.2}, Val Acc: {4:>7.2%}, Time: {5} {6}'
print(msg.format(total_batch, loss_train, acc_train, loss_val, acc_val, time_dif, improved_str))
session.run(model.optim, feed_dict=feed_dict) # 运行优化
total_batch += 1
if total_batch - last_improved > require_improvement:
# 验证集正确率长期不提升，提前结束训练
print("No optimization for a long time, auto-stopping...")
flag = True
break
if flag:
break
def train():
print("configuring tensorboard and saver")
tensorboard_dir="tensorboard/textcnn"
if not os.path.exists(tensorboard_dir):
os.makedirs(tensorboard_dir)
tf.summary.scalar("loss",model.loss)
tf.summary.scalar("accuracy",model.acc)
merged_summary=tf.summary.merge_all()
writer=tf.summary.FileWriter(tensorboard_dir)
saver=tf.train.Saver()
if not os.path.exists(save_dir):
os.makedirs(save_dir)
print("loading training and validation data..")
start_time=time.time()
x_train,y_train=process_file(train_dir,word_to_id,cat_to_id,config.seq_length)
x_val,y_val=process_file(val_dir,word_to_id,cat_to_id,config.seq_length)
time_dif=get_time_dif(start_time)
print("time usage: ",time_dif)
#创建session
session=tf.Session()
session.run(tf.global_variables_initializer())
writer.add_graph(session.graph)
print("training and evaluate...")
start_time=time.time()
total_batch=0
best_acc_val=0.0
last_improved=0
require_improvement=1000
flag=False
for epoch in range(config.num_epochs):
print("epoch: ",epoch+1)
batch_train=batch_iter(x_train,y_train,config.batch_size)
for x_batch,y_batch in batch_train:
feed_dict=feed_data(x_batch,y_batch,config.dropout_keep_prob)
if(total_batch%config.save_per_batch==0):
s=session.run(merged_summary,feed_dict=feed_dict)
writer.add_summary(s,total_batch)
if(total_batch%config.print_per_batch==0):
feed_dict[model.keep_prob]=1.0
loss_train,acc_train=session.run([model.loss,model.acc],feed_dict=feed_dict)
loss_val,acc_val=evaluate(
session,x_val,y_val
)
# 保存最好的结果
if acc_val>best_acc_val:
best_acc_val=acc_val
last_improved=0
saver.save(sess=session,save_path=save_path)
improved_str="*"
else:
improved_str=""
time_dif=get_time_dif(start_time)
#msg="iter: {0:>6}, train loss: {1:6.2},train acc:{2:7.2},"+"val loss: {3:>6.2},val acc: {4:>7.2%},time:{5},{6}"
msg = 'Iter: {0:>6}, Train Loss: {1:>6.2}, Train Acc: {2:>7.2%},' \
+ ' Val Loss: {3:>6.2}, Val Acc: {4:>7.2%}, Time: {5} {6}'
print(msg.format(total_batch, loss_train, acc_train, loss_val, acc_val, time_dif, improved_str))
session.run(model.optim,feed_dict=feed_dict)
total_batch+=1
if total_batch-last_improved>require_improvement:
print("no optimization for a long time, auto-stopping...")
flag=True
break
if flag:
break
