如何从tensorflow 2 summary writer读取数据

Question

我在从 tensorflow 摘要编写器中读取数据时遇到问题。

我正在使用 tensorflow 网站示例中的编写器：https://www.tensorflow.org/tensorboard/migrate

import tensorflow as tf
from tensorboard.backend.event_processing.event_accumulator import EventAccumulator

writer = tf.summary.create_file_writer("/tmp/mylogs/eager")

# write to summary writer
with writer.as_default():
  for step in range(100):
    # other model code would go here
    tf.summary.scalar("my_metric", 0.5, step=step)
    writer.flush()

# read from summary writer
event_acc = EventAccumulator("/tmp/mylogs/eager")
event_acc.Reload()
event_acc.Tags()

产量：

 'distributions': [],
 'graph': False,
 'histograms': [],
 'images': [],
 'meta_graph': False,
 'run_metadata': [],
 'scalars': [],
 'tensors': ['my_metric']}```

如果我尝试获取张量数据：

import pandas as pd
pd.DataFrame(event_acc.Tensors('my_metric'))

我没有得到正确的值：

wall_time   step    tensor_proto
0   1.590743e+09    3   dtype: DT_FLOAT\ntensor_shape {\n}\ntensor_con...
1   1.590743e+09    20  dtype: DT_FLOAT\ntensor_shape {\n}\ntensor_con...
2   1.590743e+09    24  dtype: DT_FLOAT\ntensor_shape {\n}\ntensor_con...
3   1.590743e+09    32  dtype: DT_FLOAT\ntensor_shape {\n}\ntensor_con...
...

如何获取实际的汇总数据（对于 100 步，每步应该为 0.5）？

这是一个包含以上代码的 colab notebook：https://colab.research.google.com/drive/1RlgZrGD_vY-YcOBLF_sEPelmtVuygkqz?usp=sharing

Answer 1

您需要转换事件累加器中的张量值，存储为TensorProto messages, into arrays, which you can do with tf.make_ndarray:

pd.DataFrame([(w, s, tf.make_ndarray(t)) for w, s, t in event_acc.Tensors('my_metric')],
             columns=['wall_time', 'step', 'tensor'])

Answer 2

为了避免部分步骤，我建议：

event_acc = EventAccumulator("/tmp/mylogs/eager", size_guidance={'tensors': 0})

如何从tensorflow 2 summary writer读取数据

How to read data from tensorflow 2 summary writer

python

tensorflow

tensorboard

tensorflow2.0