My favorites | Sign in
Project Home Wiki Issues Source
READ-ONLY: This project has been archived. For more information see this post.
Search
for
DevelopmentTackerDocument  
This page gives the detail of each and every step of the development process.
Phase-Implementation, Phase-Deploy, Phase-Support
Updated Feb 4, 2010 by just...@gmail.com

Development Tracker Document

Date : 12th March 2009

Written By Deepak Shakya

Following software have been installed for setting up the server environment. The server environment is Linux based.

1. Python 2.5

2. Natural Language Toolkit

3. NumPy - Numerical Python adds a fast and sophisticated array facility to the Python language. NumPy is the most recent and most actively supported package.

4. MatPlotLib - Matplotlib is a pure python plotting library with the goal of making publication quality plots using a syntax familiar to matlab users. The library uses numpy for handling large data sets and supports a variety of output backends

5. Prover9 - Prover9 is an automated theorem prover for first-order and equational logic, and Mace4 searches for finite models and counter examples.

6. PyBrain - yBrain is a modular Machine Learning Library for Python. It's goal is to offer flexible, easy-to-use yet still powerful algorithms for Machine Learning Tasks and a variety of predefined environments to test and compare your algorithms.

7. VPython -


Python 2.5

One can compile the source code for Python 2.5. For debian systems, type following command

sudo apt-get install python



Natural Language Toolkit (NLTK)

Obtain the zip file from here. Extract and install the toolkit by running setup.py



To get the NLTK data -

  • python -m nltk.downloader

  • Select wordnet and words to fetch the list of words in the toolkit

NumPy -

sudo apt-get install python-numpy -- debian based systems

Download and compile the tar.gz for others



MatPlotLib

sudo apt-get install python-matplotlib -- debian based systems

Download and compile the tar.gz for others.



Prover9

sudo apt-get install prover9 -- debian based systems

Download the tar.gz here and compile it.



PyBrain

Here you would find the steps to install the PyBrain library.

We preferred the zip version to compile the code. Got some error with g++

Here is the error looks like --

distutils.errors.CompileError: command 'g++' failed with exit status 1



Exact details of the error are --

`arac/src/c/layers/common.c: In function 'void make_layer(Layer, int, int)':

arac/src/c/layers/common.c:12: error: ‘malloc’ was not declared in this scope

arac/src/c/layers/common.c: In function 'Layer
make_layer(int, int)':

arac/src/c/layers/common.c:40: error: 'malloc' was not declared in this scope`



So, got the patch here. Its in Japanese and was not getting translated. But, i guess the solution was pretty simple so my very good friends knightsamar phrased it out... ;) Kudos to you. Here is what he did.

  • open this file in the extracted folder - arac/src/c/common.h
  • insert #include<stdlib.h> in the file and save it.
  • sudo python setup.py build and then sudo python setup.py install
The same steps are available at that Japanese site.



VPython

This one is trickier. You have to get some other stuff before you can compile this one.

--GtkGLExt

  • GLX - you will need this. So download and extract it. Copy the folder into the gtkglext folder. And copy the header files in the top level directory
  • Install libgl1-mesa-glx, mesa-dev, and mesa-common-dev
  • Then install this sudo apt-get install freeglut3 freeglut3-dev.
  • Now the make create the problem. Anyways, commented the line 41 in following file - gtkglext-1.2.0/gdk/x11/gdkglglxext.h.
  • make install - if you're lucky enough, you might get the stuff done.
--GtkGLExtmm
 `No package 'gtkmm-2.4' found
 No package 'gdkmm-2.4' found
 No package 'pangomm-1.4' found`
 
So we installed the gtkmm (libgtkmm-2.4-dev) - (sudo apt-get install libgtkmm-2.4-dev) package which i guess has gdkmm and pango as their dependent packages. Anyways, the package got compiled. Kudos to the knightsamar once again...;)





  • Finally the ./configure was successful.
  • Another problem with the make command.
  • Figured it out that it needs some boosting dosages (libraries like libboost-python-dev, libboost-signals-dev, and libboost-thread-dev). It might say for other dependencies as well. Go ahead and install them too. May be they boost the performance of the system.
  • So, make is done.
  • And make install too. Hurray...

Day Close Note -

  • Still have to compile vpython
  • Checked the nltk for synonyms - chapter 2 of the book (Got synonyms, definition, examples)


Powered by Google Project Hosting