Skip to content
This repository has been archived by the owner on Nov 17, 2023. It is now read-only.

Different default values for NDArray and Symbol in random.multinomial #16103

Closed
igolan opened this issue Sep 5, 2019 · 2 comments
Closed

Different default values for NDArray and Symbol in random.multinomial #16103

igolan opened this issue Sep 5, 2019 · 2 comments
Labels

Comments

@igolan
Copy link
Contributor

igolan commented Sep 5, 2019

Description

mxnet.symbol.random.multinomial
and
mxnet.ndarray.random.multinomial
have different default values for the argument get_prob.
That confusing, specifically when using hybrid blocks, and calling
F.random.multinomial
without specifying a value for get_prob.

Signatures:

mxnet.ndarray.random.multinomial(data, shape=_Null, get_prob=False, out=None, dtype='int32', **kwargs)

vs.

mxnet.symbol.random.multinomial(data, shape=_Null, get_prob=True, dtype='int32', **kwargs)

Environment info (Required)

N/A

Package used (Python/R/Scala/Julia):
MXNET1.5 python API

Build info (Required if built from source)

N/A

Error Message:

N/A

Minimum reproducible example

N/A

Steps to reproduce

https://mxnet.incubator.apache.org/versions/master/api/python/ndarray/random.html#mxnet.ndarray.random.multinomial
and
https://mxnet.incubator.apache.org/versions/master/api/python/symbol/symbol.html#mxnet.symbol.random.multinomial

What have you tried to solve it?

N/A

@mxnet-label-bot
Copy link
Contributor

Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and suggest some labels so that the appropriate MXNet community members can help resolve it.
Here are my recommended label(s): Bug

@sxjscience sxjscience added the Bug label Sep 5, 2019
@apeforest
Copy link
Contributor

Since this will impact backward compatibility and there is a plan to consolidate symbol and ndarray modules in 2.0, we will address this issue in MXNet 2.0: #9686

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
Projects
None yet
Development

No branches or pull requests

4 participants