Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Sunhwa's Meetups post #409

Open
wants to merge 2 commits into
base: gh-pages
Choose a base branch
from
Open

Sunhwa's Meetups post #409

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
48e818c
created a new file for meet-ups.
sunhwap Apr 26, 2014
f89228a
created a new post
sunhwap Apr 26, 2014
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
106 changes: 106 additions & 0 deletions _posts/2014-04-26-sunhwameetuppost.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,106 @@
---
layout: post
author: sunhwap
title: meet-ups
date: 2014-04-26
---

##[Meetup 1: TriPython: Scikit-learn](https://plus.google.com/114167959694685098579/posts/Nh3qxLw8YhB)

I attended TriPython meet-up with Brittany on the 27th of February at 7:00 pm. This meet-up was two hours long
and was taking place at WebAssign on NCSU campus. About 30 people were attending the talk and seriously listening
to the talk. Since we were late, we missed the beginning part of the talk from the presenter whose name is Tim Hopper.
He is data prediction analyst for startup parse.ly. However, based on the rest of presentation, the talk was about
introducing Scikit-learn library (http://scikit-learn.org/stable/). and what Scikit can do. I can see that Scikit
is very useful for analyzing any large data because Scikit transforms large data into matrix and estimates how good of
a fit the model score is. Full information about TriPython is available on the website at http://tripython.org.

Scikit is described as simple and efficient tools for data mining and data analysis because Scikit affords programmers
to avoid many pages of long lines of codes. He demonstrated with examples. One example has to do with transforming four
dimensional data into one dimension through building pipelines, which play a role of calculating estimators in intermediate
steps. In other words, pipelines do all the steps of transformation in chains to have one final estimator for the model fit
of a predictive statistical model. This example reminded me of one of my nephews who are getting master degree in applied
statistics because he said that he uses Python extensively for data analysis when I asked him if he uses SAS for data analysis.
Most of people at the meeting seem to have not only programming knowledge but also statistics. The presenter said that he took
two linear algebra courses as an undergraduate and two advanced linear algebra courses as a graduate. Even though he seems to
obtain higher level of mathematical knowledge, he said that he should have taken more math courses and told the audience to take
more math courses. The audience including me laughed.

Another example has to do with building predictive models in text data from 20 different news groups. One can use Scikit to
build mathematical model to organize unstructured text data that are fetched from 20 different groups. And then it transforms
string of texts into matrix and estimates how good of the model fits. So Scikit is useful for testing good predictors for a
predictive model. This talk was very informative and interesting as another form of learning process, and made me see why
it is necessary to attend different meet-ups. If I want to learn anything new, attending technology meet-ups would be a way.
I look forward to going different meet-ups in near future.

##[Meetups 2 : Pearl Hacks](https://plus.google.com/114167959694685098579/posts/EYnzZqESUWo)

I attended Pearl Hacks on the 22th of March for the second meet-up, and it was held on UNC campus.
This meet-up began with Key note address by Poornima Vijayashanker, who is founding engineer at Mint.com and Femgineer.
Her talk was interesting and inspiring for me. When she took her first programing course, she earned only 15 out of 100
on her first computer programming test and dropped the course for fear of getting poor final grade for the course.
However, instead of being discouraged, she buried herself in a library and spent many hours practicing programming and
ended up earning a solid A when she took the same course again. Then, she decided to do double major in electrical
engineering and computer science at duke, but her decision to do double major was too late to finish both degrees in 4 years.
This meant that she had to go an additional year to finish both degrees. However, since her parents were in a financial
hardship at that time, she could not afford to attend college for an additional year. However, instead of giving up earning
an extra degree, she endured the heavy course loads by working very hard and successfully finished both degrees. She worked
in Silicon Valley as a software engineer and has been an avid supporter and a mentor to the organizers for meet-ups like
Pearl Hacks for females who have much interest in technology.

I went to Pearl hacks not for hackathon but mainly for one of the skill sessions since I could not stay overnight due to
a health reason. Brittany and I were interested in learning visual representation of google data. But she had to go to
a group meeting session since she was going to participate in the hackathon, which required overnight stay. We were assuming
that the group meeting would be brief. So we agreed to go to her group meeting session first and then go to the skill session.
However, the meeting that we assumed to take 10 to 15 minutes turned out to be an hour long. So we already missed the bulk of
the skill session, and since Brittany’s group wanted to go to the Android session, I followed them along for it. Even though
I felt disappointed at not being able to make the skill session for which I came, the brainstorming session for group projects
provided me with another learning experience.

I watched the group brainstorming session for those who do not have group members for hackathon. As mentors started to bring up
ideas and suggestions for topics for the student participants led by one of the organizers, students pitched in with their ideas.
It gave me a glimpse of how program developers would come up with a project and process of narrowing down the topics and
preferences among group members. This brainstorming helped me to see why sites like Github are of use for a cooperative project,
and I was glad about the fact that my Python course exposed me to such a site even though it was confusing since I forget the
details easily. After the brainstorming session, Brittany’s group and I went to the android app development skill room. We were
waiting to have workshop but even after 30 minutes waiting, the instructor did not do anything. It turned out that presentation
was already done, and each group was already working on some sort of task that probably taught by the instructor since we were
one hour late. As a result, I was accidently deprived of learning any skill that I eagerly expected to learn through this event.
But I was glad to come since it was another form of learning experience for me, and this event motivated me to plan to go to events
like this in the near future.

##[Meetup 3: TriPython: Brython talk](https://plus.google.com/114167959694685098579/posts/EYnzZqESUWo)

I went to another meet-up of TryPython on the 27th of March. This meet-up took place at Bull City Coworking in Durham. The talk
was about Brython. The presenter, whose name is Francios Dion, is the Python developer at Inmar, founder of Dion Research. His
interest lies in educational angle, mentoring and helping others to learn python. In his words, “Brython is Python in your browser”,
which means one does not have to write Javascript. Brython is designed to replace JavaScript as the scripting language for the web.
It is a python 3 implementation for your web browser. I was wondering why it is called Brython. After hearing “Brython is Python
in your browser”, I immediately assumed that Br was prefixed in front of ython because of the word browser. However, my guess about
the name was wrong. It is called Brython because it was created by Brenton Pierre Quentel, and Br is Welsh for Breton.

I was wondering why Python in the browser. The presenter said that it is because that is the only way to interact with your command
line in some situations. So how does it work? Brython engine was written in Javascript (py2js.js), and it scans the HTML for Python
source and builds tree presenting Python codes and then translate them to Javascript. Brython is pretty new, and as such is changing
rapidly. But, you can use it pretty much anywhere without server side code. Brython is fast enough and has support for mobile device.
Brython itself seems to be completely client side, but whether that will be enough really depends on the code you write. It is not a
full blown Python interpreter and does not have the libraries.

One thing that drew more of my attention was his use of a familiar word, Github when he introduced Bitbucket near the end of the
presentation. Bitbucket is a web-based hosting service site for projects like Github. Out of curiosity, I searched Bitbucket to see the
difference between Github and Bitbucket. Both GitHub and Bitbucket support public and private repositories. Since GitHub doesn't allow
for external authentication, each member of your team will have to set up a new account at GitHub. Bitbucket, on the other hand, allows
you to log in with Twitter, Google, Facebook, OpenID, and GitHub credentials. But there are four times more Github users than Bitbucket
users. I was wondering why Github attracts far more users. It seems that Github works great for open-source sharing “social coding”,
and the number of collaborators is usually positively correlated with whether the project is open-source or not. Bitbucket users are
those people who would like to keep their non-open-source projects since Bitbucket provides unlimited free private repositories, while
Github charges for private repo based on the pricing model. So If I need repositories for any non-open source project, it is good to
know that I can turn to Bitbucket.

## Reflection

Overall, it was beneficial for me to attend three different meet-ups because two Python meets-up exposed me to useful libraries
while Perl meet-ups provided skill sessions. Attending the meet-ups has planted a seed in me to be inclined to attend more meet-ups like Pearl where
you get to learn different technology skills and hone your coding skills. Attending meet-ups is a great way to be exposed to different
and useful resources for anyone who has much interest in programming. I am glad that going to meet-ups was necessry for this class
since it provided me with another form of learning.
Loading