TensorFlow-如何设置外部过程的梯度(PY_FUNC)?
-
16-10-2019 - |
题
我正在计算和优化外部进程上使用的某些变量,但是我遇到了“无梯度”的错误。
代码的大量简化(未测试)版本,但是您可以得到这个想法:
def external_process (myvar):
subprocess.call("process.sh", myvar)
with open('result.json', 'r') as f:
result = json.load(data, f)
return np.array(result["result"])
myvar = tf.Variable(1.0, dtype = 'float32', trainable = True)
loss = tf.reduce_sum( tf.py_func(external_process, [myvar], [tf.float32])[0] )
optimizer = tf.train.AdamOptimizer(0.05)
train_step = optimizer.minimize(loss)
sess.run(train_step)
我看到了这个讨论,但我不完全理解: https://github.com/tensorflow/tensorflow/issues/1095
谢谢!
解决方案 2
正确的答案是:“外部过程是无法区分的(除非您知道每个细节,在这种情况下是不可能的),因此该问题应作为加强学习问题而面对的”
其他提示
这里 https://www.tensorflow.org/versions/r0.9/api_docs/python/framework.html (搜索 gradient_override_map
)是一个例子 gradient_override_map
:
@tf.RegisterGradient("CustomSquare")
def _custom_square_grad(op, grad):
# ...
with tf.Graph().as_default() as g:
c = tf.constant(5.0)
s_1 = tf.square(c) # Uses the default gradient for tf.square.
with g.gradient_override_map({"Square": "CustomSquare"}):
s_2 = tf.square(s_2) # Uses _custom_square_grad to compute the
# gradient of s_2.
因此,可能的解决方案可能是:
@tf.RegisterGradient("ExternalGradient")
def _custom_external_grad(unused_op, grad):
# I don't know yet how to compute a gradient
# From Tensorflow documentation:
return grad, tf.neg(grad)
def external_process (myvar):
subprocess.call("process.sh", myvar)
with open('result.json', 'r') as f:
result = json.load(data, f)
return np.array(result["result"])
myvar = tf.Variable(1.0, dtype = 'float32', trainable = True)
g = tf.get_default_graph()
with g.gradient_override_map({"PyFunc": "ExternalGradient"}):
external_data = tf.py_func(external_process, [myvar], [tf.float32])[0]
loss = tf.reduce_sum(external_data)
optimizer = tf.train.AdamOptimizer(0.05)
train_step = optimizer.minimize(loss)
sess.run(train_step)