Control flow operators do not work in imperative (NDArray) API

[samskalicky]

Description

The example for cond operator does not work for NDArray API from the website ([https://mxnet.incubator.apache.org/api/python/ndarray/contrib.html#mxnet.ndarray.contrib.cond])

a, b = mx.nd.array([1]), mx.nd.array([2])
pred = a * b < 5
then_func = lambda a, b: (a + 5) * (b + 5)
else_func = lambda a, b: (a - 5) * (b - 5)
outputs = mx.nd.contrib.cond(pred, then_func, else_func)
outputs[0]

When running this with v1.3.x here is the output:

>>> a, b = mx.nd.array([1]), mx.nd.array([2])
>>> pred = a * b < 5
>>> then_func = lambda a, b: (a + 5) * (b + 5)
>>> else_func = lambda a, b: (a - 5) * (b - 5)
>>> outputs = mx.nd.contrib.cond(pred, then_func, else_func)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/lib/python2.7/site-packages/mxnet/ndarray/contrib.py", line 460, in cond
    return then_func()
TypeError: <lambda>() takes exactly 2 arguments (0 given)

The symbolic API example does work for the cond operator. Is it supposed to work for for the NDArray API? If so, then this is a bug. If not, then the documentation needs to be fixed since its not supported in the NDArray API.

[samskalicky]

@zheng-da What is your take on this? Is this supposed to work, or does the documentation need to be removed for cond (and other control flow operators) in the NDArray API?

[piyushghai]

@mxnet-label-bot [Question, NDArray]

[wkcn]

It seems the example is wrong.

The following example works.

import mxnet as mx
a, b = mx.nd.array([1]), mx.nd.array([2])
pred = a * b < 5
then_func = lambda : (a + 5) * (b + 5)
else_func = lambda : (a - 5) * (b - 5)
outputs = mx.nd.contrib.cond(pred, then_func, else_func)
print (outputs[0])

[ThomasDelteil]

@samskalicky can you please close the issue? @sandeep-krishnamurthy can you please close the issue the PR has been merged