具有numpy数组的张量板

Question

有人可以举例说明如何使用张量板可视化numpy数组值吗？

这里有一个相关的问题，我真的不明白。 Tensorboard logging non-tensor (numpy) information (AUC)

例如， 如果我有

for i in range(100):
    foo = np.random.rand(3,2)

如何使用tensorboard跟踪foo的分布100次迭代？有人可以给出代码示例吗？ 谢谢。

Answer 1

对于简单值（标量），您可以使用此配方

summary_writer = tf.train.SummaryWriter(FLAGS.logdir)
summary = tf.Summary()
summary.value.add(tag=tagname, simple_value=value)
summary_writer.add_summary(summary, global_step)
summary_writer.flush()

就使用数组而言，也许您可​​以在序列中添加6个值，即

for value in foo:
  summary.value.add(tag=tagname, simple_value=value)

Answer 2

另一种（最简单的）方法就是使用占位符。首先，您可以为numpy数组形状创建占位符。

# Some place holders for summary
summary_reward = tf.placeholder(tf.float32, shape=(), name="reward")
tf.summary.scalar("reward", summary_reward)

然后，只需使用feed_dict调用session.run合并的摘要。

# Summary
summ = tf.summary.merge_all()
...
s = sess.run(summ, feed_dict={summary_reward: reward})
writer.add_summary(s, i)

Answer 3

如果您通过pip install tensorboard-pytorch安装this package，它就会变得尽可能简单：

import numpy as np
from tensorboardX import SummaryWriter
writer = SummaryWriter()
for i in range(50):
    writer.add_histogram("moving_gauss", np.random.normal(i, i, 1000), i, bins="auto")
writer.close()

将在runs目录中生成相应的直方图数据：

Answer 4

找到一种解决方法，创建一个变量并将numpy数组的值赋给变量，使用tensorboard来跟踪变量

mysummary_writer = tf.train.SummaryWriter("./tmp/test/")
a = tf.Variable(tf.zeros([3,2]), name="a")
sum1 = tf.histogram_summary("nparray1", a)
summary_op = tf.merge_all_summaries()
sess = tf.Session()

sess.run(tf.initialize_all_variables())

for ii in range(10):
    foo = np.random.rand(3, 2)
    assign_op = a.assign(foo)
    summary, _ = sess.run([summary_op, assign_op])
    mysummary_writer.add_summary(tf.Summary.FromString(summary), global_step=ii)
    mysummary_writer.flush()

Answer 5

sess = tf.Session()
writer = tf.summary.FileWriter('tensorboard_test')
var = tf.Variable(0.0,trainable=False,name='loss')
sess.run(var.initializer)
summary_op = tf.summary.scalar('scalar1',var)

for value in array:    
    sess.run(var.assign(value))
    summary = sess.run(summary_op)
    writer.add_summary(summary,i)

虽然有效，但速度很慢。

Answer 6

你可以定义一个这样的函数（取自gyglim的要点）：

def add_histogram(writer, tag, values, step, bins=1000):
    """
    Logs the histogram of a list/vector of values.
    From: https://gist.github.com/gyglim/1f8dfb1b5c82627ae3efcfbbadb9f514
    """

    # Create histogram using numpy
    counts, bin_edges = np.histogram(values, bins=bins)

    # Fill fields of histogram proto
    hist = tf.HistogramProto()
    hist.min = float(np.min(values))
    hist.max = float(np.max(values))
    hist.num = int(np.prod(values.shape))
    hist.sum = float(np.sum(values))
    hist.sum_squares = float(np.sum(values ** 2))

    # Requires equal number as bins, where the first goes from -DBL_MAX to bin_edges[1]
    # See https://github.com/tensorflow/tensorflow/blob/master/tensorflow/core/framework/summary.proto#L30
    # Therefore we drop the start of the first bin
    bin_edges = bin_edges[1:]

    # Add bin edges and counts
    for edge in bin_edges:
        hist.bucket_limit.append(edge)
    for c in counts:
        hist.bucket.append(c)

    # Create and write Summary
    summary = tf.Summary(value=[tf.Summary.Value(tag=tag, histo=hist)])
    writer.add_summary(summary, step)

然后像这样添加到摘要作者：

add_histogram(summary_writer, "Histogram_Name", your_numpy_array, step)