mkdir mygit
cd mygit
git init
git add .
git commit -m 'initial commit'

Checkout repo

cd mygit
git clone url myrepo

Create branch

git branch -a 
git checkout -b branchtoworkon origin/existingbranch

or

git checkout existingbranch
git branch branchtoworkon
git checkout branchtoworkon

Push branch

git push origin branchtoworkon

Create tag

git tag -l
git tag 1.0

Push tag

git push origin 1.0

Delete branch/tag

git tag -d 1.0
git branch -d branchtoworkon
git push origin :branchtoworkon

So.

$ sudo apt-get install python-virtualenv
$ virtualenv -p /usr/bin/python2.4 /home/cawanpink/python2.4
$ source /home/cawanpink/python2.4/bin/activate
(python2.4)$ deactivate

That’s

• install
• create the Python 2.4 virtual environment
• activate the environment
• to get out of the environment

What’s data mining?

Well, basically it’s a process to get a knowledge (eg. if customers buy diapers, most likely they will buy beers too) from your data (imagine spreadsheets filled with numbers and strings). You may have thousands of records but how do you make sense of it? You mine those data, and get a pattern. From there you’ll get a knowledge. So from this you can put beers next to diapers in supermarkets to increase sales. That’s the general idea.

There’s a 90% chance that the data you got is not cleaned aka there’s missing values, some data are not consistent (eg. in sex can only have F or M, there’s a value Q?), and some data are just plain rubbish. The data need to go through preprocessing stage. For missing values, you have to filled them up, either using mean or median values, whichever is best for your data. The same goes to inconsistent data. This is where working with experts in the domain you’re working in is very important. You don’t want to remove what you think is rubbish but it actually means something.

Then we have data discretization process where you reduce that huge amount of data but they still carry the same value. Afterwards we normalize the data. After this is done, then the data is ready for the modelling exercise.

Classification

One of the benefit you get from data mining is classification. It’s where you can predict if you have this data (eg. male, married, doesn’t have children, have regularly purchase beer for the past year) whether beer purchase is likely. So it will classify to something like this purchase_beer equals to 0 (no) or 1 (yes). In order for the model to predict 0 or 1, the model itself has to be trained with alot of data. Train it until it reaches the accuracy we want, above 90% is good. A trained model with very high accuracy is going to be an asset as you can feed it data and it will spit out what we want.

The most difficult part in doing this for me is the data preprocessing part. You have to have quality cleaned data to produce quality results. You have to carefully select which kind of data is relevant to your goal (eg. would you want to include one’s job as one of the attributes considered for the diapers-beers example?) which is why having domain experts is important. They also have to determine how each data should affect the result (eg. job probably affects 10% but marital status affects 80% towards diapers-beers purchase).

I just love AI

There are just so many preprocessing methods & data mining hybrid techniques already been researched by academicians. I’m just so overwhelmed by the amount of technical papers on this. They probably not so much IT savvy like us, the implementors as they call it, but they definitely have the brain on that part of the world. We just have to scour through this massive database and get it to run on our apps. Well, should you need it that is. Coz processing thousands of data can take hours or days, some months, depending on your machine spec.

Open data

While doing data mining assignment last 2 weeks, I became frustrated with the unavailability of data in Malaysia. Sad. Maybe it’s still difficult for us to see it now, I’m already imagining all the stuff we can do with those data like in JPA, MOHR, MOH, MOHE – fuhhhh I’m all shuddery now. Even in OSCC, those training feedback forms and MyGOSSCON feedback forms – hmm whatever happened to those?

There are concerns of exposing private data I guess. Well, if you ask me make the data anonymous, as I couldn’t care less who got promoted last year. I only want the ’spec’ of that person who got promoted – age, sex, location, department, salary, is he respected by peers, does he drive, etc, that kind of stuff, you get the idea.

Hmm.. this will take another 10 years to realise, I think.

MyMeeting is also on Launchpad that makes use of Bazaar. We wanted to find a way how to send changes to both its main repo (using SVN) and Launchpad.

So we’ve been using SVN for MyMeeting hosted at OSCC. A typical way for us would be like this.

$ svn co https://svn.oscc.org.my/mymeeting/trunk trunk
$ cd trunk
(hack hack hack...)
$ svn status #see our changes
$ svn ci -m 'added feature ABC' #commit to SVN repository

To use Bazaar to work with SVN repo, you have to install bzr and bzr-svn. Excellent doc on bzr-szv is here.

$ sudo apt-get bzr bzr-svn

SVN-like

Working with Bazaar, the way would be something like this.

$ mkdir dev
$ bzr init-repo --default-rich-root dev
$ cd dev
$ bzr co https://svn.oscc.org.my/mymeeting/trunk trunk
$ cd trunk
(hack hack hack...)
$ bzr update #get changes done by others
$ bzr ci -m 'added form for feature ABC' #commit to SVN repository
$ bzr push lp:mymeeting #push to Launchpad, only have to provide location once
(hack hack hack...)
$ bzr update #get changes done by others
$ bzr ci -m 'added list for feature ABC' #commit to SVN repository
$ bzr push

Decentralized Bazaar way

If we were to take advantage of Bazaar’s decentralised way of doing it (so you can work offline, for example), it’s like this.

$ mkdir dev
$ bzr init-repo --default-rich-root dev
$ cd dev
$ bzr co https://svn.oscc.org.my/mymeeting/trunk trunk #our copy of trunk
$ bzr branch trunk working #make a local branch to hack on
$ cd working
(hack hack hack...working offline)
$ bzr ci -m 'added form for feature ABC' #commit to local branch
(hack hack hack...working offline)
$ bzr ci -m 'added list for feature ABC' #commit to local branch
 
(when you get your connection back)
$ cd ../trunk
$ bzr update #get changes done by others to our copy of trunk 
$ cd ../working
$ bzr pull #pull the changes to our local branch
$ bzr status #see our changes
$ cd ../trunk
$ bzr merge ../working
$ bzr ci -m 'added feature ABC'

Personally I like the centralised approach because it’s similar to SVN. Local branch is great feature if I have to do my work offline sometimes. And while I can pick Bazaar from now on, the rest of the team doesn’t have to switch tool. That’s great!

1st prize: RM5,000 cash
2nd prize: RM3,000 cash
3rd prize: RM1,500 cash

Anyway yes it’s open for all, and they have additional prize for best team for students. If you’re in IPTA/IPTS, you’ll probably end up RM2k richer.

More information here.

That’s Martha, Nicholas’ daughter (of OSCC). She won the OLPC hands down during FOSS.my 2008, beating all other grown ups in a little game for OLPC giveaway. I think she’s the youngest audience in the hall. Nice trick Nic, I know you want the OLPC for geek self :p!

FOSS.my 2009 - little Martha holding her OLPC

The translation page.

I’ve been doing it for a while now, and it’s kinda fun(ny) translating English to Malay. I only armed myself with wiki and kamus dewan (go go online dictionary!). One can only wonders how to translate “Headers and Footers” or “Plug and Play” (thanks Eric for the idea haha!). Sometimes I just settle with sentences that can make sense, it doesn’t have to be translated literally, or word by word. So far I haven’t faced any major difficulties yet, only minor setbacks, like I couldn’t find the translation for some very techie word in malay but that’s ok. I’ll leave it to the reviewers muahahaha!

No seriously, the hardest part would be the ones that involves the shorcuts. That’s tough. Apart from it has to be consistent, it also has to make sense too. I think the best way to do it to list down all available shortcuts in one app, assign all those letters and then translate it. But someone has to be in charge of this.

I noticed that I can download the .POT file. It’s so much easier to translate using say, Poedit, then using the web based. We’re using Poedit for MyMeeting translation by the way – i’s kicking ass as of now. But I couldn’t find a way to upload back the translation file to Launchpad. Maybe the settings? I didn’t have time to investigate further. Web based is fine for me for now.

One thing that I like about Launchpad is it’s displaying the context of the strings to be translated. Example, if you were to be given “Page”, how would you determine whether it’s “Laman” as in the website page or “Muka Surat” as in pages from a book? Come to think of it, we can use laman for both but you get the general idea right?

Ok have to break fast for the last day of Ramadhan now. We’re having the usual rendang and ketupat for tomorrow’s eid! There’s also ketupat palas, which anyone of you doesn’t know of, I’ll post a pic or two later. I’m in Penang now, and I expect to have lots of laksa. Yum!

P/S: Btw, Header = Pengepala, Footer = Pengaki. I had my laughs, now it’s your turn.

It’s a 3 page folder, where you open it like a book and then you can open it again to the right. Inside it should have 18 small samples of tiles of various colors. So a complete folder should weigh around 7kg (I can’t hold it for too long, it’s like carrying a 7 year old around in one hand ). It’s done purely using Inkscape (so called Adobe Illustrator equivalent).

MSA Worldwide Sdn Bhd sample folder - 18 tiles

His client in Dubai, UAE is happy so hey I’m going to celebrate it tonight! There’s one more folder need to be done but smaller size.

I just hope, although it’s very unlikely, that the printer will accept SVG file instead of AI for a change! Now wouldn’t that make a world a better place? Freedom of choice. Either that or a very high resolution PNG.

1. Read a file (*.txt) and determine frequency of each characters and total bits.
2. Build a Huffman tree based on previous frequency table. Each node keeps a character & its frequency, path (eg. 011, 0011, 01110 etc), vertex flag (whether it’s a left or right child) and visit flag (whether it has been visited or not – for node searching later). Note though, not every node keeps a character, only leaf nodes do (Huffman tree properties). Sorting algorithm to build priority queue to work with when building the tree is selection sorting algorithm.
3. Follow every root-to-leaf path to get the bits for each character. This script uses breadth-first search algorithm to look for nodes inside the tree.
4. Write compression information in a new file (compressed one) but this script only write the bits sequence into the file.

The file here: huffman.cpp (C++)

Some screenshots: