Floating point exception in mxnet.ndarray.InstanceNorm

[leeyeetonn]

Description

(A clear and concise description of what the bug is.)
mxnet.ndarray.InstanceNorm has floating point exception when the given data's batch = 0 or channel = 0. Please see the provided code snippets for example.

Error Message

(Paste the complete error message. Please also include stack trace by setting environment variable DMLC_LOG_STACK_TRACE_DEPTH=10 before running your script.)

Floating point exception (core dumped)

To Reproduce

(If you developed your own code, please provide a short script that reproduces the error. For existing examples, please provide link.)

import mxnet
import numpy as np

input = mxnet.nd.array(np.random.rand(0,1,1)) # batch = 0
gamma = mxnet.nd.array(np.random.rand(1))
beta = mxnet.nd.array(np.random.rand(1))
mxnet.ndarray.InstanceNorm(input, gamma, beta)

import mxnet
import numpy as np

input = mxnet.nd.array(np.random.rand(1,0,1)) # channel = 0
gamma = mxnet.nd.array(np.random.rand(0))
beta = mxnet.nd.array(np.random.rand(0))
mxnet.ndarray.InstanceNorm(input, gamma, beta)

Steps to reproduce

(Paste the commands you ran that produced the error.)

1. run the provided code in python interpreter or as a script

What have you tried to solve it?

Environment

We recommend using our script for collecting the diagnositc information. Run the following command and paste the outputs below:

curl --retry 10 -s https://github.com/dmlc/gluon-nlp/master/tools/diagnose.py | python

# paste outputs here

Got 404 when trying to get the script.

Some environment information:

• OS: ubuntu 18.04
• Python: 3.7.6
• pip: 20.0.2
• numpy: 1.18.5
• mxnet: 1.6.0

[szha]

Confirmed that this issue also happens in numpy branch. Since instance norm was using a legacy operator interface, I will refactor it to use the new operator interface and register an npx namespace alias.

from mxnet import np, npx

npx.set_np()

input = np.random.rand(0,1,1) # batch = 0
gamma = np.random.rand(1)
beta = np.random.rand(1)
npx.instance_norm(input, gamma, beta)

from mxnet import np, npx

npx.set_np()

input = np.random.rand(1,0,1) # channel = 0
gamma = np.random.rand(1)
beta = np.random.rand(1)
npx.instance_norm(input, gamma, beta)