You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2018/06/28 19:44:44 UTC
[GitHub] larroy opened a new issue #11468: Inconsistent / wrong output from
sum
larroy opened a new issue #11468: Inconsistent / wrong output from sum
URL: https://github.com/apache/incubator-mxnet/issues/11468
Followup of problems found during https://github.com/apache/incubator-mxnet/issues/9853
## Description
Output inconsistent across runs and wrong when summing two arrays, also observed for other operators like bmod. But only after running sum.
Couldn't reproduce with the pip installable package for 1.3.0
Can't reproduce without debug flags
```
output:
[[2.5355392 3.314111 2.1423295 ]
[4.1634665 3.7890441 3.1607347 ]
[3.2863414 2.973648 3.8752291 ]
[1.704005 0.11532945 4.315019 ]]
data:
[array([[0.51929279, 0.71588298, 0.70266935],
[0.98779978, 0.36271363, 0.0987926 ],
[0.95248205, 0.98068743, 0.01229028],
[0.28883612, 0.06484356, 0.32346107]]), array([[0.5040616 , 0.86607602, 0.35991496],
[0.79391669, 0.85658257, 0.76548548],
[0.77795309, 0.66432023, 0.96573467],
[0.471723 , 0.02524295, 0.99788945]])]
```
## Environment info (Required)
```
Ubuntu 16.04 c5d.18xlarge
```
```
----------Python Info----------
Version : 3.5.2
Compiler : GCC 5.4.0 20160609
Build : ('default', 'Nov 23 2017 16:37:01')
Arch : ('64bit', 'ELF')
------------Pip Info-----------
Version : 10.0.1
Directory : /home/piotr/mxnet/mxnet_py3/lib/python3.5/site-packages/pip
----------MXNet Info-----------
[19:29:52] ../src/engine/engine.cc:55: MXNet start using engine: NaiveEngine
Version : 1.3.0
Directory : /home/piotr/mxnet/python/mxnet
Hashtag not found. Not installed from pre-built package.
----------System Info----------
Platform : Linux-4.4.0-1060-aws-x86_64-with-Ubuntu-16.04-xenial
system : Linux
node : ip-172-31-44-40
release : 4.4.0-1060-aws
version : #69-Ubuntu SMP Sun May 20 13:42:07 UTC 2018
----------Hardware Info----------
machine : x86_64
processor : x86_64
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 72
On-line CPU(s) list: 0-71
Thread(s) per core: 2
Core(s) per socket: 18
Socket(s): 2
NUMA node(s): 2
Vendor ID: GenuineIntel
CPU family: 6
Model: 85
Model name: Intel(R) Xeon(R) Platinum 8124M CPU @ 3.00GHz
Stepping: 4
CPU MHz: 3000.000
BogoMIPS: 6000.00
Hypervisor vendor: KVM
Virtualization type: full
L1d cache: 32K
L1i cache: 32K
L2 cache: 1024K
L3 cache: 25344K
NUMA node0 CPU(s): 0-17,36-53
NUMA node1 CPU(s): 18-35,54-71
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc arc
h_perfmon rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave a
vx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single kaiser fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx avx512f rdseed adx smap clflushop
t clwb avx512cd xsaveopt xsavec xgetbv1 ida arat
----------Network Test----------
Setting timeout: 10
Timing for Gluon Tutorial(cn): https://zh.gluon.ai, DNS: 0.1978 sec, LOAD: 1.0500 sec.
Timing for Gluon Tutorial(en): http://gluon.mxnet.io, DNS: 0.1359 sec, LOAD: 0.3103 sec.
Timing for FashionMNIST: https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/dataset/fashion-mnist/train-labels-idx1-ubyte.gz, DNS: 0.0391 sec, LOAD: 0.2770 se
c.
Timing for PYPI: https://pypi.python.org/pypi/pip, DNS: 0.0033 sec, LOAD: 0.4341 sec.
Timing for MXNet: https://github.com/apache/incubator-mxnet, DNS: 0.0007 sec, LOAD: 0.7057 sec.
Timing for Conda: https://repo.continuum.io/pkgs/free/, DNS: 0.0141 sec, LOAD: 0.3944 sec.
```
Package used (Python/R/Scala/Julia):
python
## Build info (Required if built from source)
Compiler (gcc/clang/mingw/visual studio):
```
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-linux-gnu/5/lto-wrapper
Target: x86_64-linux-gnu
Configured with: ../src/configure -v --with-pkgversion='Ubuntu 5.5.0-12ubuntu1~16.04' --with-bugurl=file:///usr/share/doc/gcc-5/README.Bugs --enable-languages=c,ada,c++,java,go,d,fortran,objc,obj-c++ --prefix=/usr --program-suffix=-5 --enable-shared --enable-linker-build-id --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --libdir=/usr/lib --enable-nls --with-sysroot=/ --enable-clocale=gnu --enable-libstdcxx-debug --enable-libstdcxx-time=yes --with-default-libstdcxx-abi=new --enable-gnu-unique-object --disable-vtable-verify --enable-libmpx --enable-plugin --with-system-zlib --disable-browser-plugin --enable-java-awt=gtk --enable-gtk-cairo --with-java-home=/usr/lib/jvm/java-1.5.0-gcj-5-amd64/jre --enable-java-home --with-jvm-root-dir=/usr/lib/jvm/java-1.5.0-gcj-5-amd64 --with-jvm-jar-dir=/usr/lib/jvm-exports/java-1.5.0-gcj-5-amd64 --with-arch-directory=amd64 --with-ecj-jar=/usr/share/java/eclipse-ecj.jar --enable-objc-gc --enable-multiarch --disable-werror --with-arch-32=i686 --with-abi=m64 --with-multilib-list=m32,m64,mx32 --enable-multilib --with-tune=generic --enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu --target=x86_64-linux-gnu
Thread model: posix
gcc version 5.5.0 20171010 (Ubuntu 5.5.0-12ubuntu1~16.04)
```
MXNet commit hash:
1b69e4bd92dc3c2a73a310993cd21e6119d52c0c
Build config:
```
#!/bin/bash
set -e
set -x
renice -n 19 -p $$
mkdir -p build && cd build
cmake -DUSE_CPP_PACKAGE=ON -DUSE_CUDA=OFF -DUSE_OPENMP=OFF -DUSE_OPENCV=ON -DCMAKE_BUILD_TYPE=Debug -GNinja ..
ninja -v
cd ..
if [ ! -d mxnet_py3 ]; then
virtualenv -p `which python3` mxnet_py3
fi
source mxnet_py3/bin/activate
cd python
pip install -e .
cd ..
pip install opencv-python
pip install ipython
pip install matplotlib
```
## Error Message:
## Minimum reproducible example
```
#!/usr/bin/env python
# -*- coding: utf-8 -*-
"""Description"""
__author__ = 'Pedro Larroy'
__version__ = '0.1'
import os
import sys
import subprocess
import numpy as np
import mxnet as mx
import random
def gen_data(dummy):
ndim = np.random.randint(1, 6)
shape = np.random.randint(1, 6, size=(ndim,))
print("gen shape {}".format(shape))
return [np.random.random(shape), np.random.random(shape)]
def main():
np.random.seed(0)
mx.random.seed(0)
random.seed(0)
sample_num = 200
for i in range(sample_num):
d = gen_data(i)
a = mx.sym.Variable('a')
b = mx.sym.Variable('b')
c = a + b
y = c.bind(mx.cpu(), args={'a': mx.nd.array(d[0]), 'b': mx.nd.array(d[1])})
y.forward(is_train=True)
y = y.outputs[0].asnumpy()
print("output:")
print(y)
print("data:")
print(d)
return 1
if __name__ == '__main__':
sys.exit(main())
```
## Steps to reproduce
(Paste the commands you ran that produced the error.)
```
python repro.py > 1
python repro.py > 2
diff 1 2
```
## What have you tried to solve it?
1. Set MXNET_ENGINE_TYPE=NaiveEngine problem still persists
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services