def train(self):
if self.conf.reload_epoch > 0:
self.reload(self.conf.reload_epoch)
train_reader = H5DataLoader(self.conf.data_dir+self.conf.train_data)
valid_reader = H5DataLoader(self.conf.data_dir+self.conf.valid_data)
for epoch_num in range(self.conf.max_epoch):
if epoch_num % self.conf.test_step == 1:
inputs, annotations = valid_reader.next_batch(self.conf.batch)
feed_dict = {self.inputs: inputs,
self.annotations: annotations}
loss, summary = self.sess.run(
[self.loss_op, self.valid_summary], feed_dict=feed_dict)
self.save_summary(summary, epoch_num+self.conf.reload_epoch)
print('----testing loss', loss)
elif epoch_num % self.conf.summary_step == 1:
inputs, annotations = train_reader.next_batch(self.conf.batch)
feed_dict = {self.inputs: inputs,
self.annotations: annotations}
loss, _, summary = self.sess.run(
[self.loss_op, self.train_op, self.train_summary],
feed_dict=feed_dict)
self.save_summary(summary, epoch_num+self.conf.reload_epoch)
else:
inputs, annotations = train_reader.next_batch(self.conf.batch)
feed_dict = {self.inputs: inputs,
self.annotations: annotations}
loss, _ = self.sess.run(
[self.loss_op, self.train_op], feed_dict=feed_dict)
print('----training loss', loss)
if epoch_num % self.conf.save_step == 1:
self.save(epoch_num+self.conf.reload_epoch)
def train(self):
if self.conf.reload_step > 0:
self.reload(self.conf.reload_step)
train_reader = H5DataLoader(self.conf.data_dir + self.conf.train_data)
valid_reader = H5DataLoader(self.conf.data_dir + self.conf.valid_data)
for epoch_num in range(self.conf.max_step + 1):
if epoch_num and epoch_num % self.conf.test_interval == 0:
inputs, labels = valid_reader.next_batch(self.conf.batch)
feed_dict = {self.inputs: inputs, self.labels: labels}
loss, summary = self.sess.run(
[self.loss_op, self.valid_summary], feed_dict=feed_dict)
self.save_summary(summary, epoch_num + self.conf.reload_step)
print('----testing loss', loss)
if epoch_num and epoch_num % self.conf.summary_interval == 0:
inputs, labels = train_reader.next_batch(self.conf.batch)
feed_dict = {self.inputs: inputs, self.labels: labels}
loss, _, summary = self.sess.run(
[self.loss_op, self.train_op, self.train_summary],
feed_dict=feed_dict)
self.save_summary(summary, epoch_num + self.conf.reload_step)
else:
inputs, labels = train_reader.next_batch(self.conf.batch)
feed_dict = {self.inputs: inputs, self.labels: labels}
loss, _ = self.sess.run([self.loss_op, self.train_op],
feed_dict=feed_dict)
print("Epoch", epoch_num, '----training loss', loss)
if epoch_num and epoch_num % self.conf.save_interval == 0:
self.save(epoch_num + self.conf.reload_step)
def train(self):
# 有时可以从以训练好的model开始训练
if self.conf.reload_epoch > 0:
self.reload(self.conf.reload_epoch)
# 读取数据
train_reader = H5DataLoader(self.conf.data_dir+self.conf.train_data)
valid_reader = H5DataLoader(self.conf.data_dir+self.conf.valid_data)
# 记录loss
valid_loss_list = []
train_loss_list = []
for epoch_num in range(self.conf.max_epoch):
if epoch_num % self.conf.test_step == 1:
inputs, annotations = valid_reader.next_batch(self.conf.batch)
feed_dict = {self.inputs: inputs, self.annotations: annotations}
loss, summary = self.sess.run([self.loss_op, self.valid_summary], feed_dict=feed_dict)
self.save_summary(summary, epoch_num)
print(epoch_num, '----testing loss', loss)
print(epoch_num)
# 记录验证集上的loss
valid_loss_list.append(loss)
np.save(self.conf.record_dir+"valid_loss.npy",np.array(valid_loss_list))
elif epoch_num % self.conf.summary_step == 1:
inputs, annotations = train_reader.next_batch(self.conf.batch)
feed_dict = {self.inputs: inputs, self.annotations: annotations}
loss, _, summary = self.sess.run([self.loss_op, self.train_op, self.train_summary], feed_dict=feed_dict)
self.save_summary(summary, epoch_num)
print(epoch_num)
# 记录训练集上的loss
train_loss_list.append(loss)
np.save(self.conf.record_dir+"train_loss.npy",np.array(train_loss_list))
else:
inputs, annotations = train_reader.next_batch(self.conf.batch)
feed_dict = {self.inputs: inputs, self.annotations: annotations}
loss,_ = self.sess.run([self.loss_op, self.train_op], feed_dict=feed_dict)
print(epoch_num)
# 记录训练集上的loss
train_loss_list.append(loss)
np.save(self.conf.record_dir+"train_loss.npy",np.array(train_loss_list))
if epoch_num % self.conf.save_step == 1:
self.save(epoch_num)
def test(self):
print('---->testing ', self.conf.test_epoch)
if self.conf.test_epoch > 0:
self.reload(self.conf.test_epoch)
else:
print("please set a reasonable test_epoch")
return
test_reader = H5DataLoader(
self.conf.data_dir+self.conf.test_data, False)
self.sess.run(tf.local_variables_initializer())
count = 0
losses = []
accuracies = []
m_ious = []
while True:
inputs, annotations = test_reader.next_batch(self.conf.batch)
if inputs.shape[0] < self.conf.batch:
break
feed_dict = {self.inputs: inputs, self.annotations: annotations}
loss, accuracy, m_iou, _ = self.sess.run(
[self.loss_op, self.accuracy_op, self.m_iou, self.miou_op],
feed_dict=feed_dict)
print('values----->', loss, accuracy, m_iou)
count += 1
losses.append(loss)
accuracies.append(accuracy)
m_ious.append(m_iou)
print('Loss: ', np.mean(losses))
print('Accuracy: ', np.mean(accuracies))
print('M_iou: ', m_ious[-1])
def test(self,model_i):
print('---->testing ', model_i)
# 加载模型
if model_i > 0:
self.reload(model_i)
else:
print("please set a reasonable test_epoch")
return
# 读取数据，注意是False，代表不是在训练
valid_reader = H5DataLoader(self.conf.data_dir+self.conf.valid_data,False)
self.sess.run(tf.local_variables_initializer())
# 记录测试参数
losses = []
accuracies = []
m_ious = []
while True:
inputs, annotations = valid_reader.next_batch(self.conf.batch)
# 终止条件：当取出的batch不够个数了就break
if inputs.shape[0] < self.conf.batch:
break
feed_dict = {self.inputs: inputs, self.annotations: annotations}
loss, accuracy, m_iou, _ = self.sess.run([self.loss_op, self.accuracy_op, self.m_iou, self.miou_op], feed_dict=feed_dict)
print('values----->', loss, accuracy, m_iou)
losses.append(loss)
accuracies.append(accuracy)
m_ious.append(m_iou)
# 其实是每一个batch上计算一次指标，最后求均值
return np.mean(losses),np.mean(accuracies),m_ious[-1]
def train(self):
if self.conf.reload_epoch > 0:
self.reload(self.conf.reload_epoch)
data_reader = H5DataLoader('../../Data/data/celeba_train_test.h5')
epoch_num = 1
epoch = 0
while epoch < self.conf.max_epoch:
if epoch_num % self.conf.test_step == 1:
inputs = np.zeros(self.input_shape)
feed_dict = {self.inputs: inputs, self.valid: True}
summary = self.sess.run(self.valid_summary,
feed_dict=feed_dict)
self.save_summary(summary, epoch_num)
elif epoch_num % self.conf.summary_step == 1:
targets = data_reader.next_batch(self.conf.batch)
feed_dict = {self.inputs: targets, self.valid: False}
_, summary = self.sess.run(
[self.train_opg, self.train_summary], feed_dict=feed_dict)
self.save_summary(summary, epoch_num)
else:
targets = data_reader.next_batch(self.conf.batch)
feed_dict = {self.inputs: targets, self.valid: False}
rec_loss_l, kl_loss, _ = self.sess.run(
[self.rec_loss_l, self.kl_loss, self.train_opg],
feed_dict=feed_dict)
print('epoch = ', epoch, '----training loss l=', rec_loss_l,
', kl= ', kl_loss)
if epoch_num % self.conf.save_step == 0:
self.save(epoch_num)
epoch = data_reader.epoch
epoch_num += 1
def predict(self):
print('---->predicting ', self.conf.test_epoch)
if self.conf.test_epoch > 0:
self.reload(self.conf.test_epoch)
else:
print("please set a reasonable test_epoch")
return
test_reader = H5DataLoader(
self.conf.data_dir+self.conf.test_data, False)
self.sess.run(tf.local_variables_initializer())
predictions = []
losses = []
accuracies = []
m_ious = []
count=0
while True:
inputs, annotations = test_reader.next_batch(self.conf.batch)
if inputs.shape[0] < self.conf.batch:
break
feed_dict = {self.inputs: inputs, self.annotations: annotations}
loss, accuracy, m_iou, _ = self.sess.run(
[self.loss_op, self.accuracy_op, self.m_iou, self.miou_op],
feed_dict=feed_dict)
print('values----->', loss, accuracy, m_iou)
losses.append(loss)
accuracies.append(accuracy)
m_ious.append(m_iou)
predictions.append(self.sess.run(
self.decoded_predictions, feed_dict=feed_dict))
print('----->saving predictions')
for index, prediction in enumerate(predictions):
for i in range(prediction.shape[0]):
imsave(prediction[i], self.conf.sample_dir +
str(index*prediction.shape[0]+i)+'.png')
return np.mean(losses),np.mean(accuracies),m_ious[-1]
def predict(self):
print('---->predicting ', self.conf.test_epoch)
if self.conf.test_epoch > 0:
self.reload(self.conf.test_epoch)
else:
print("please set a reasonable test_epoch")
return
# 读取数据
test_reader = H5DataLoader(self.conf.data_dir + self.conf.test_data,
False)
self.sess.run(tf.local_variables_initializer())
predictions = []
losses = []
accuracies = []
m_ious = []
while True:
inputs, annotations = test_reader.next_batch(self.conf.batch)
# 终止条件
if inputs.shape[0] < self.conf.batch:
break
feed_dict = {self.inputs: inputs, self.annotations: annotations}
loss, accuracy, m_iou, _ = self.sess.run(
[self.loss_op, self.accuracy_op, self.m_iou, self.miou_op],
feed_dict=feed_dict)
print('values----->', loss, accuracy, m_iou)
# 记录指标
losses.append(loss)
accuracies.append(accuracy)
m_ious.append(m_iou)
# 记录预测值
predictions.append(
self.sess.run(self.decoded_predictions, feed_dict=feed_dict))
print('----->saving predictions')
print(np.shape(predictions))
num = 0
for index, prediction in enumerate(predictions):
# 下面的程序用于输出一通道的预测值，测试时需要观察的
#print(prediction.shape)
#print(index)
#np.save("pred",np.array(prediction))
# 把一通道的预测值保存为三通道图片，这是自己写的函数
for i in range(prediction.shape[0]):
np.save(self.conf.sample_dir + "pred" + str(num) + ".npy",
prediction[i])
num += 1
imsave(
prediction[i], self.conf.sample_dir +
str(index * prediction.shape[0] + i) + '.png')
# 验证和测试的时候，指标都是返回的全体上的均值
return np.mean(losses), np.mean(accuracies), m_ious[-1]
def test(self, model_i):
print('---->testing ', model_i)
if model_i > 0:
self.reload(model_i)
else:
print("please set a reasonable test_epoch")
return
valid_reader = H5DataLoader(self.conf.data_dir + self.conf.valid_data,
False)
self.sess.run(tf.local_variables_initializer())
losses = []
accuracies = []
m_ious = []
dices = []
count = 0
while True:
inputs, annotations = valid_reader.next_batch(self.conf.batchsize)
if inputs.shape[0] < self.conf.batch:
break
feed_dict = {
self.inputs: inputs,
self.annotations: annotations,
self.is_train: False
}
loss, accuracy, m_iou, _ = self.sess.run(
[self.loss_op, self.accuracy_op, self.m_iou, self.miou_op],
feed_dict=feed_dict)
print(count)
print('values----->', loss, accuracy, m_iou)
losses.append(loss)
accuracies.append(accuracy)
m_ious.append(m_iou)
out, gt = self.sess.run([self.out, self.gt], feed_dict=feed_dict)
if self.conf.class_num == 2:
tp = np.sum(out * gt)
fenmu = np.sum(out) + np.sum(gt) + 0.000001
dice = 2 * tp / fenmu
dices.append(dice)
print('dice----->', dice)
count += 1
if count == self.conf.valid_num:
break
return np.mean(losses), np.mean(accuracies), m_ious[-1], np.mean(dices)
def test(self):
print('---->testing ', self.conf.test_step)
if self.conf.test_step > 0:
self.reload(self.conf.test_step)
else:
print("please set a reasonable test_step")
return
test_reader = H5DataLoader(self.conf.data_dir + self.conf.test_data,
False)
accuracies = []
while True:
inputs, labels = test_reader.next_batch(self.conf.batch)
if inputs is None or inputs.shape[0] < self.conf.batch:
break
feed_dict = {self.inputs: inputs, self.labels: labels}
accur = self.sess.run(self.accuracy_op, feed_dict=feed_dict)
accuracies.append(accur)
print('accuracy is ', sum(accuracies) / len(accuracies))
def predict(self):
self.reload(self.conf.test_epoch)
test_reader = H5DataLoader(self.conf.data_dir + self.conf.test_data,
False)
self.sess.run(tf.local_variables_initializer())
predictions = []
net_predictions = []
losses = []
dices = []
accuracies = []
count = 0
while True:
inputs, annotations = test_reader.next_batch(self.conf.batchsize)
if inputs.shape[0] < self.conf.batch:
break
feed_dict = {
self.inputs: inputs,
self.annotations: annotations,
self.is_train: False
}
predictions.append(
self.sess.run(
self.predictions,
feed_dict=feed_dict)) # <-------decoded_predictions
net_predictions.append(
self.sess.run(self.decoded_net_pred, feed_dict=feed_dict))
count += 1
if count == self.conf.test_num:
break
num = 0
for index, prediction in enumerate(net_predictions):
for i in range(prediction.shape[0]):
num += 1
imsave(
prediction[i], self.conf.sample_net_dir +
str(index * prediction.shape[0] + i) + '.png')
return np.mean(losses), np.mean(accuracies), np.mean(dices)
def output_train(self):
print('---->output', self.conf.test_epoch)
if self.conf.test_epoch > 0:
self.reload(self.conf.test_epoch)
else:
print("please set a reasonable test_epoch")
return
data_reader = H5DataLoader('../../Data/data/celeba_train_test.h5')
targets = data_reader.next_batch(self.conf.batch)
feed_dict = {self.inputs: targets, self.valid: False}
predictions, super_predictions = self.sess.run(
[self.predictions, self.super_predictions], feed_dict=feed_dict)
super_predictions = np.array(super_predictions)
predictions = np.array(predictions)
for i in range(predictions.shape[0]):
imsave(self.conf.sample_dir + str(i) + '_o.png',
np.reshape(targets[i], (128, 128, 3)))
#imsave(self.conf.sample_dir + str(i) + 'l.png', np.reshape(predictions[i], (128,128,3)))
imsave(self.conf.sample_dir + str(i) + '_h.png',
np.reshape(super_predictions[i], (128, 128, 3)))
def predict(self):
print('---->predicting ', self.conf.test_epoch)
if self.conf.test_epoch > 0:
self.reload(self.conf.test_epoch)
else:
print("please set a reasonable test_epoch")
return
test_reader = H5DataLoader(
self.conf.data_dir+self.conf.test_data, False)
predictions = []
while True:
inputs, annotations = test_reader.next_batch(self.conf.batch)
if inputs.shape[0] < self.conf.batch:
break
feed_dict = {self.inputs: inputs, self.annotations: annotations}
predictions.append(self.sess.run(
self.decoded_predictions, feed_dict=feed_dict))
print('----->saving predictions')
for index, prediction in enumerate(predictions):
for i in range(prediction.shape[0]):
imsave(prediction[i], self.conf.sample_dir +
str(index*prediction.shape[0]+i)+'.png')
def train(self):
# 用于记录summary
self.train_summary = self.config_summary('train')
self.valid_summary = self.config_summary('valid')
# 有时可以从已训练好的model开始训练
if self.conf.reload_epoch > 0:
self.reload(self.conf.reload_epoch)
# 读取数据
train_reader = H5DataLoader(self.conf.data_dir + self.conf.train_data)
valid_reader = H5DataLoader(self.conf.data_dir + self.conf.valid_data)
# 记录loss
valid_loss_list = []
train_loss_list = []
# 记录accuracy
train_acc_list = []
valid_acc_list = []
# 记录m_iou
train_miou_list = []
valid_miou_list = []
# 初始化局部变量是为了保存训练中的 miou， 因为这是个局部变量
self.sess.run(tf.local_variables_initializer())
# 开始训练
for epoch_num in range(self.conf.max_epoch):
# 训练到test_step，在验证集上进行一次验证
if epoch_num % self.conf.test_step == 1:
inputs, annotations = valid_reader.next_batch(
self.conf.batchsize)
feed_dict = {
self.inputs: inputs,
self.annotations: annotations,
self.is_train: False
}
#loss, summary = self.sess.run([self.loss_op, self.valid_summary], feed_dict=feed_dict)
loss, accuracy, m_iou, _ = self.sess.run(
[self.loss_op, self.accuracy_op, self.m_iou, self.miou_op],
feed_dict=feed_dict)
#self.save_summary(summary, epoch_num)
print(epoch_num, '----valid loss', loss)
# 记录验证集上的loss
valid_loss_list.append(loss)
np.save(self.conf.record_dir + "valid_loss.npy",
np.array(valid_loss_list))
# 记录验证集上的acc
valid_acc_list.append(accuracy)
np.save(self.conf.record_dir + "valid_acc.npy",
np.array(valid_acc_list))
# 记录验证集上的miou
valid_miou_list.append(m_iou)
np.save(self.conf.record_dir + "valid_miou.npy",
np.array(valid_miou_list))
################################### 还是要做训练的呀 #######################################
inputs, annotations = train_reader.next_batch(
self.conf.batchsize)
feed_dict = {
self.inputs: inputs,
self.annotations: annotations,
self.is_train: True
}
haha, loss, accuracy, m_iou, _ = self.sess.run(
[
self.train_op, self.loss_op, self.accuracy_op,
self.m_iou, self.miou_op
],
feed_dict=feed_dict)
print(epoch_num, '----train loss', loss)
# 记录训练集上的loss
train_loss_list.append(loss)
np.save(self.conf.record_dir + "train_loss.npy",
np.array(train_loss_list))
# 记录训练集上的acc
train_acc_list.append(accuracy)
np.save(self.conf.record_dir + "train_acc.npy",
np.array(train_acc_list))
# 记录训练集上的miou
train_miou_list.append(m_iou)
np.save(self.conf.record_dir + "train_miou.npy",
np.array(train_miou_list))
elif epoch_num % self.conf.summary_step == 1:
inputs, annotations = train_reader.next_batch(
self.conf.batchsize)
feed_dict = {
self.inputs: inputs,
self.annotations: annotations,
self.is_train: False
}
#loss, _, summary = self.sess.run([self.loss_op, self.train_op, self.train_summary], feed_dict=feed_dict)
#self.save_summary(summary, epoch_num)
#print(epoch_num)
# 记录训练集上的loss
#train_loss_list.append(loss)
#np.save(self.conf.record_dir+"train_loss.npy",np.array(train_loss_list))
else:
inputs, annotations = train_reader.next_batch(
self.conf.batchsize)
feed_dict = {
self.inputs: inputs,
self.annotations: annotations,
self.is_train: True
}
loss, _ = self.sess.run([self.loss_op, self.train_op],
feed_dict=feed_dict)
print(epoch_num)
# 保存模型
if epoch_num % self.conf.save_step == 1:
self.save(epoch_num)
def predict(self):
print('---->predicting ', self.conf.test_epoch)
if self.conf.test_epoch > 0:
self.reload(self.conf.test_epoch)
else:
print("please set a reasonable test_epoch")
return
test_reader = H5DataLoader(self.conf.data_dir + self.conf.test_data,
False)
self.sess.run(tf.local_variables_initializer())
predictions = []
net_predictions = []
outputs = []
probabilitys = []
losses = []
accuracies = []
m_ious = []
rate_list = []
befores = []
afters = []
maps = []
start_maps = []
count = 0
while True:
inputs, annotations = test_reader.next_batch(self.conf.batchsize)
if inputs.shape[0] < self.conf.batch:
break
feed_dict = {
self.inputs: inputs,
self.annotations: annotations,
self.is_train: False
}
loss, accuracy, m_iou, _ = self.sess.run(
[self.loss_op, self.accuracy_op, self.m_iou, self.miou_op],
feed_dict=feed_dict)
print('values----->', loss, accuracy, m_iou)
losses.append(loss)
accuracies.append(accuracy)
m_ious.append(m_iou)
predictions.append(
self.sess.run(self.decoded_predictions, feed_dict=feed_dict))
net_predictions.append(
self.sess.run(self.decoded_net_pred, feed_dict=feed_dict))
outputs.append(self.sess.run(self.outputs, feed_dict=feed_dict))
count += 1
if count == self.conf.test_num:
break
print('----->saving outputs')
print(np.shape(probabilitys))
np.save(self.conf.sample_dir + "outputs" + ".npy", np.array(outputs))
print('----->saving predictions')
print(np.shape(predictions))
num = 0
for index, prediction in enumerate(predictions):
for i in range(prediction.shape[0]):
np.save(self.conf.sample_dir + "pred" + str(num) + ".npy",
prediction[i])
num += 1
imsave(
prediction[i], self.conf.sample_dir +
str(index * prediction.shape[0] + i) + '.png')
print('----->saving net_predictions')
print(np.shape(net_predictions))
num = 0
for index, prediction in enumerate(net_predictions):
for i in range(prediction.shape[0]):
np.save(self.conf.sample_dir + "netpred" + str(num) + ".npy",
prediction[i])
num += 1
imsave(
prediction[i], self.conf.sample_dir +
str(index * prediction.shape[0] + i) + 'net.png')
return np.mean(losses), np.mean(accuracies), m_ious[-1]
def train(self):
self.train_summary = self.config_summary('train')
self.valid_summary = self.config_summary('valid')
if self.conf.reload_epoch > 0:
self.reload(self.conf.reload_epoch)
train_reader = H5DataLoader(self.conf.data_dir + self.conf.train_data)
valid_reader = H5DataLoader(self.conf.data_dir + self.conf.valid_data)
valid_loss_list = []
train_loss_list = []
train_acc_list = []
valid_acc_list = []
train_miou_list = []
valid_miou_list = []
self.sess.run(tf.local_variables_initializer())
for epoch_num in range(self.conf.max_epoch):
if epoch_num % self.conf.test_step == 1:
inputs, annotations = valid_reader.next_batch(
self.conf.batchsize)
feed_dict = {
self.inputs: inputs,
self.annotations: annotations,
self.is_train: False
}
#loss, summary = self.sess.run([self.loss_op, self.valid_summary], feed_dict=feed_dict)
loss, accuracy, m_iou, _ = self.sess.run(
[self.loss_op, self.accuracy_op, self.m_iou, self.miou_op],
feed_dict=feed_dict)
#self.save_summary(summary, epoch_num)
print(epoch_num, '----valid loss', loss)
# loss
valid_loss_list.append(loss)
np.save(self.conf.record_dir + "valid_loss.npy",
np.array(valid_loss_list))
# acc
valid_acc_list.append(accuracy)
np.save(self.conf.record_dir + "valid_acc.npy",
np.array(valid_acc_list))
# miou
valid_miou_list.append(m_iou)
np.save(self.conf.record_dir + "valid_miou.npy",
np.array(valid_miou_list))
#########################################################################
inputs, annotations = train_reader.next_batch(
self.conf.batchsize)
feed_dict = {
self.inputs: inputs,
self.annotations: annotations,
self.is_train: True
}
haha, loss, accuracy, m_iou, _ = self.sess.run(
[
self.train_op, self.loss_op, self.accuracy_op,
self.m_iou, self.miou_op
],
feed_dict=feed_dict)
print(epoch_num, '----train loss', loss)
# loss
train_loss_list.append(loss)
np.save(self.conf.record_dir + "train_loss.npy",
np.array(train_loss_list))
# acc
train_acc_list.append(accuracy)
np.save(self.conf.record_dir + "train_acc.npy",
np.array(train_acc_list))
# miou
train_miou_list.append(m_iou)
np.save(self.conf.record_dir + "train_miou.npy",
np.array(train_miou_list))
elif epoch_num % self.conf.summary_step == 1:
inputs, annotations = train_reader.next_batch(
self.conf.batchsize)
feed_dict = {
self.inputs: inputs,
self.annotations: annotations,
self.is_train: False
}
#loss, _, summary = self.sess.run([self.loss_op, self.train_op, self.train_summary], feed_dict=feed_dict)
#self.save_summary(summary, epoch_num)
#print(epoch_num)
#train_loss_list.append(loss)
#np.save(self.conf.record_dir+"train_loss.npy",np.array(train_loss_list))
else:
inputs, annotations = train_reader.next_batch(
self.conf.batchsize)
feed_dict = {
self.inputs: inputs,
self.annotations: annotations,
self.is_train: True
}
loss, _ = self.sess.run([self.loss_op, self.train_op],
feed_dict=feed_dict)
print(epoch_num)
if epoch_num % self.conf.save_step == 1:
self.save(epoch_num)