2

在TensorFlow Distributions（现为Probability ）的参考论文中，提到了 TensorFlow可用于构造和对象，即：VariableBijectorTransformedDistribution

import tensorflow as tf
import tensorflow_probability as tfp
tfd = tfp.distributions

tf.enable_eager_execution()

shift = tf.Variable(1., dtype=tf.float32)
myBij = tfp.bijectors.Affine(shift=shift)

# Normal distribution centered in zero, then shifted to 1 using the bijection
myDistr = tfd.TransformedDistribution(
            distribution=tfd.Normal(loc=0., scale=1.),
            bijector=myBij,
            name="test")

# 2 samples of a normal centered at 1:
y = myDistr.sample(2)
# 2 samples of a normal centered at 0, obtained using inverse transform of myBij:
x = myBij.inverse(y)

我现在想修改 shift 变量（比如说，我可能会计算一些似然函数的梯度作为 shift 的函数并更新它的值）所以我这样做

shift.assign(2.)
gx = myBij.forward(x)

我希望如此gx=y+1，但我看到了gx=y……确实，myBij.shift仍然评估为1.

如果我尝试直接修改双射器，即：

myBij.shift.assign(2.)

我明白了

AttributeError: 'tensorflow.python.framework.ops.EagerTensor' object has no attribute 'assign'

计算梯度也不能按预期工作：

with tf.GradientTape() as tape:
    gx = myBij.forward(x)
grad = tape.gradient(gx, shift)

Yields None，以及脚本结束时的此异常：

Exception ignored in: <bound method GradientTape.__del__ of <tensorflow.python.eager.backprop.GradientTape object at 0x7f529c4702e8>>
Traceback (most recent call last):
File "~/.local/lib/python3.6/site-packages/tensorflow/python/eager/backprop.py", line 765, in __del__
AttributeError: 'NoneType' object has no attribute 'context'

我在这里想念什么？

编辑：我让它与图形/会话一起工作，所以似乎急切执行存在问题......

注意：我有 tensorflow 版本 1.12.0 和 tensorflow_probability 版本 0.5.0

4

1 回答 1

1

如果您使用的是 Eager 模式，则需要从变量向前重新计算所有内容。最好在函数中捕获此逻辑；

import tensorflow as tf
import tensorflow_probability as tfp
tfd = tfp.distributions

tf.enable_eager_execution()

shift = tf.Variable(1., dtype=tf.float32)
def f():
  myBij = tfp.bijectors.Affine(shift=shift)

  # Normal distribution centered in zero, then shifted to 1 using the bijection
  myDistr = tfd.TransformedDistribution(
            distribution=tfd.Normal(loc=0., scale=1.),
            bijector=myBij,
            name="test")

  # 2 samples of a normal centered at 1:
  y = myDistr.sample(2)
  # 2 samples of a normal centered at 0, obtained using inverse
  # transform of myBij:
  x = myBij.inverse(y)
  return x, y
x, y = f()
shift.assign(2.)
gx, _ = f()

关于渐变，您需要将调用包装f()在GradientTape

于 2018-11-26T19:10:21.300 回答