def main():
##% matplotlib inline
params_dirname = "fit_a_line.inference.model"
x, y, y_predict = network()
feeder = fluid.DataFeeder(place=place, feed_list=[x, y])
avg_loss, main_program, test_program, startup_program = backPropagation(
y_predict, y)
exe.run(startup_program)
train_prompt = "train cost"
test_prompt = "test cost"
from paddle.utils.plot import Ploter
plot_prompt = Ploter(train_prompt, test_prompt)
step = 0
exe_test = fluid.Executor(place)
train_reader, test_reader = preProcess()
for pass_id in range(num_epochs):
for data_train in train_reader():
avg_loss_value, = exe.run(main_program,
feed=feeder.feed(data_train),
fetch_list=[avg_loss])
if step % 10 == 0: # 每10个批次记录并输出一下训练损失
plot_prompt.append(train_prompt, step, avg_loss_value[0])
plot_prompt.plot()
print("%s, Step %d, Cost %f" %
(train_prompt, step, avg_loss_value[0]))
if step % 100 == 0: # 每100批次记录并输出一下测试损失
test_metics = train_test(executor=exe_test,
program=test_program,
reader=test_reader,
fetch_list=[avg_loss.name],
feeder=feeder)
plot_prompt.append(test_prompt, step, test_metics[0])
plot_prompt.plot()
print("%s, Step %d, Cost %f" %
(test_prompt, step, test_metics[0]))
if test_metics[0] < 10.0: # 如果准确率达到要求，则停止训练
break
step += 1
if math.isnan(float(avg_loss_value[0])):
sys.exit("got NaN loss, training failed.")
# 保存训练参数到之前给定的路径中
if params_dirname is not None:
fluid.io.save_inference_model(params_dirname, ['x'],
[y_predict], exe)
exe.run(startup_program)
train_prompt = "train cost"
test_prompt = "test cost"
from paddle.utils.plot import Ploter
plot_prompt = Ploter(train_prompt, test_prompt)
step = 0
exe_test = fluid.Executor(place)
for pass_id in range(num_epochs):
for data_train in train_reader():
avg_loss_value, = exe.run(main_program,
feed=feeder.feed(data_train),
fetch_list=[avg_loss])
if step % 10 == 0: # 每10个批次记录并输出一下训练损失
plot_prompt.append(train_prompt, step, avg_loss_value[0])
#plot_prompt.plot()
print("%s, Step %d, Cost %f" %
(train_prompt, step, avg_loss_value[0]))
if step % 100 == 0: # 每100批次记录并输出一下测试损失
test_metics = train_test(executor=exe_test,
program=test_program,
reader=test_reader,
fetch_list=[avg_loss.name],
feeder=feeder)
plot_prompt.append(test_prompt, step, test_metics[0])
#plot_prompt.plot()
print("%s, Step %d, Cost %f" % (test_prompt, step, test_metics[0]))
if test_metics[0] < 10.0: # 如果准确率达到要求，则停止训练
break
def event_handler_plot(ploter_title, step, miou):
train_prompt = "Train cost"
test_prompt = "Test cost"
cost_ploter = Ploter(train_prompt, test_prompt)
cost_ploter.append(ploter_title, step, miou)
cost_ploter.plot()
loss, distance, label = loss_contrastive.forward(
Output1, Output2, Label)
avg_loss = layers.reduce_mean(loss)
sum_loss += avg_loss.numpy()[0]
print(
"Epoch number {}, step {}: Label {}; Euclidean Distance {}; Current loss {}\n"
.format(epoch, step, np.concatenate(label),
np.concatenate(distance), avg_loss.numpy()))
# 后向传播，更新参数的过程
# if distance == 0:
# continue
avg_loss.backward()
optimizer.minimize(avg_loss)
model.clear_gradients()
step += 1
train_cost.append(train_prompt, epoch, sum_loss / step)
# 保存模型目录
save_path = './checkpoint/' + Config.training_dir.split('/')[-1]
os.makedirs(save_path, exist_ok=True)
# 保存模型参数
if sum_loss < _loss:
_loss = sum_loss
save_path += '/transform_2D_epoch{}'.format(epoch)
fluid.save_dygraph(model.state_dict(), save_path)
os.makedirs('/home/aistudio/figure_and_csv', exist_ok=True)
train_cost.plot('/home/aistudio/figure_and_csv/' +
Config.training_dir.split('/')[-1] +
'_train_{}.png'.format(t))