如何编辑日志目录并保存？

Question

我有此代码，可以成功加载丢失信息。我想编辑此信息并再次保存，以便以后可以加载tensorboard进行显示，如何实现？

from tensorflow.python.summary import event_accumulator as ea

acc = ea.EventAccumulator("train")
acc.Reload()

# Print tags of contained entities, use these names to retrieve entities as below
print(acc.Tags())

xy_l2_loss = [(s.step, s.value) for s in acc.Scalars('seq2seq/loss/loss')]

print(xy_l2_loss)

Answer 1

假设您要将其存储在文件中：

import pickle
from tensorflow.python.summary import event_accumulator as ea

acc = ea.EventAccumulator("train")
acc.Reload()

# Print tags of contained entities, use these names to retrieve entities as below
print(acc.Tags())

xy_l2_loss = [(s.step, s.value) for s in acc.Scalars('seq2seq/loss/loss')]

print(xy_l2_loss)
xy_l2_loss = 'asdf'
#store data
with open('somefile.pkl', 'wb') as f:
    pickle.dump(xy_l2_loss, f)
#load data
with open('somefile.pkl', 'rb') as f:
    xy_l2_loss = pickle.load(f)