I have been trying to install graph tool for past 2 days. Have met with many errors in installing it and its dependencies (still solving). Also I have seen many errors posted by the community regarding the usage of package.
I was planning to use it for my research work since the tools seems promising. But it might take some time to change to another graph tool if I find out at a later stage that there are too many bugs in Python-graph tool and is not yet ready to relied on for a Ph.D level work.
Hence I would really like to hear from the people who have used this tool for some work before I begin.
I have been trying to install graph tool for past 2 days. Have met
with many errors in installing it and its dependencies (still
solving). Also I have seen many errors posted by the community
regarding the usage of package.
I was planning to use it for my research work since the tools seems
promising. But it might take some time to change to another graph tool
if I find out at a later stage that there are too many bugs in
Python-graph tool and is not yet ready to relied on for a Ph.D level
work.
Having problems with installation does not mean there is a bug with
graph-tool.
Being implemented in C++, and depending on other C/C++ libraries,
graph-tool requires a basic understanding and familiarity with compiling
and installing C++ libraries, sorting out dependencies, etc. For those
that do not have this familiarity, there are packages available for some
operating systems (Ubuntu, Debian, Arch, Gentoo and MacOS), that make
this task easier.
Hence I would really like to hear from the people who have used this
tool for some work before I begin.
I use it extensively in my research, and I know that others do too.
I am sorry to hear that you have installation problems with graph-tool.
It seems a bit harsh towards the hard-working creator of graph-tool to
question the feasibility of a software package for doing PhD research,
based on the fact that you lack the necessary skills to install the package.
Graph-tool is a great package for network analysis and is more efficient
than most (all?) other out-of-the-box network analysis packages that I
know.
I use it extensively for my research and at my university I have had the
package installed by the system administration on all student machines.
A number of bachelor and master students use the package from home or from
their own machines throughout my social network analysis course.
Installation requires one to follow a number of steps that should be easy
for anyone with basic knowledge of Linux and how to install software under
that operating system, requiring one to know only a little bit more than
"sudo apt-get install PackageX" under Ubuntu.
In any case, if you are doing PhD research, and need graph-tool, meaning
you somehow want to do large-scale network analysis, basic Linux skills are
likely much welcome and worth spending a few weeks learning.
Good luck,
Kind regards,
Frank
2016-06-28 7:45 GMT+02:00 vineesh vs <shopping_vineeshvs(a)outlook.com>:
Glad to know that there is a good user-base and supporting community around Python-graph tool. Never intended to question the hard-work went behind creation of such an awesome tool. Would like to apologize if it felt so.
I also had problems in compiling the latest version on my computer, but the
pre-compiled package works perfectly, and the dependencies are also easy to
install, even if you just google for "how to install cairo on ubuntu" for
example.
I'm using parts of the package for my research, combined with my code for
knot theory, and I can tell that graph-tool is flexible enough to suit your
graph theory needs. The documentation was easy to understand and has nice
examples that you can start with, although it will take some time to master
it. (I'm only mastering the parts that I need :P) Tiago even implemented
something for me from boost, that I'm thankful for.
I think you will find this package helpful and if you find something better
out there, please let me know (I did a thorough search a year ago, but
graph-tool was the best).
(I did a thorough search a year ago, but