[jira] [Comment Edited] (SINGA-274) Create Debian packages

["Moaz Reyad (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SINGA-274?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15771722#comment-15771722 ] 

Moaz Reyad edited comment on SINGA-274 at 12/23/16 3:08 AM:
------------------------------------------------------------

I created two Debian packages :

1. singa:

{code:title=Install singa (C++ and CPU only)|borderStyle=solid}
wget http://www.comp.nus.edu.sg/~moaz/singa-cpu_1.0.0_amd64.deb
apt-get install ./singa-cpu_1.0.0_amd64.deb
{code}

{code:title=Test singa installation (C++ and CPU only)|borderStyle=solid}
wget http://www.comp.nus.edu.sg/~moaz/test.cpp
g++ -std=c++11 -DUSE_GLOG -o test ./test.cpp -lglog -lsinga
{code}

{code:title=Uninstall singa (C++ and CPU only)|borderStyle=solid}
apt-get remove singa
{code}

2. singa-python

{code:title=Install singa (Python and CPU only)|borderStyle=solid}
wget http://www.comp.nus.edu.sg/~moaz/singa-python_1.0.0_amd64.deb
apt-get install ./singa-python_1.0.0_amd64.deb
{code}

{code:title=Test singa installation (Python and CPU only)|borderStyle=solid}
python
> from singa import tensor
{code}

{code:title=Uninstall singa (Python and CPU only)|borderStyle=solid}
apt-get remove singa-python
{code}

Notes:
1. The two packages are independent. If you want to use both C++ and Python, you must install both.
2. Packages are tested on Ubuntu (> 16.04), but they could work on other platforms as well.
3. The packages are binary compiled for amd64 operating systems only.

More improvements can be done for these packages, but they can be used now for learning the system and to do simple experiments and testing. This should enhance the usability of SINGA and remove the installation pain from Debian/Ubuntu end users.


was (Author: moazreyad):
I created two Debian packages :

1. singa:

{code:title=Install singa (C++ and CPU only)|borderStyle=solid}
wget http://www.comp.nus.edu.sg/~moaz/singa-cpu_1.0.0_amd64.deb
apt-get install ./singa-cpu_1.0.0_amd64.deb
{code}

{code:title=Test singa installation (C++ and CPU only)|borderStyle=solid}
wget http://www.comp.nus.edu.sg/~moaz/test.cpp
g++ -std=c++11 -DUSE_GLOG -o test ./test.cpp -lglog -lsinga
{code}

{code:title=Uninstall singa (C++ and CPU only)|borderStyle=solid}
apt-get remove singa
{code}

2. singa-python

{code:title=Install singa (Python and CPU only)|borderStyle=solid}
wget http://www.comp.nus.edu.sg/~moaz/singa-python_1.0.0_amd64.deb
apt-get install./singa-python_1.0.0_amd64.deb
{code}

{code:title=Test singa installation (Python and CPU only)|borderStyle=solid}
python
> from singa import tensor
{code}

{code:title=Uninstall singa (Python and CPU only)|borderStyle=solid}
apt-get remove singa-python
{code}

Notes:
1. The two packages are independent. If you want to use both C++ and Python, you must install both.
2. Packages are tested on Ubuntu (> 16.04), but they could work on other platforms as well.
3. The packages are binary compiled for amd64 operating systems only.

More improvements can be done for these packages, but they can be used now for learning the system and to do simple experiments and testing. This should enhance the usability of SINGA and remove the installation pain from Debian/Ubuntu end users.

> Create Debian packages
> ----------------------
>
>                 Key: SINGA-274
>                 URL: https://issues.apache.org/jira/browse/SINGA-274
>             Project: Singa
>          Issue Type: Improvement
>         Environment: Linux (Debian/Ubuntu)
>            Reporter: Moaz Reyad
>
> To make SINGA installation easier, the following Debian packages are proposed:
> singa: the core package for singa C++ on CPU
> singa-python: the pysinga, depends on singa pacakge on CPU
> singa-gpu : the gpu version of the singa C++, depends on singa and extends it with the gpu features
> singa-gpu-python: pysinga and gpu together. depends on singa-python and singa-gpu
> singa-dlaas: depends on singa-python, adds the DL as a service (Rafiki) on CPU
> singa-gpu-dlaas: depends on singa-gpu-python, adds the DL as a service (Rafiki) on GPU
> These packages can be created in Debian Launchpad (https://launchpad.net).
> Later we can proceed with sending these packages to Debian and Ubuntu official distribution.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)