Profile

Cover photo
Travis McKee
Works at New York University
Attends New York University
115 followers|197,635 views
AboutPostsPhotosVideos

Stream

Travis McKee

Shared publicly  - 
 
Ive been a pretty entrenched emacs for a while now, but I think its almost time to move on. Emacs is so slow! I wish it werent so. Might be time to bite the bullet and try sublime
Emacs
In Emacs We Trust! :-)
View community
1
Add a comment...
 
Hello all, I am working on a predictive modeling project at the moment and I am using #Python  to preprocess the data, which is complicated enough that the script to process it will be several hundred lines of code to get it into a format that could be fed into a machine learning algorithm. 

My original plan was to convert the data to be used with Weka, which is a Java machine learning platform, but it was suggested to me that if I was already this embedded in Python for the project that I might as well step it up and learn to do the machine learning modeling with Python.

So my question is which libraries in Python would provide the best tools to implement, test, and compare several different types of predictive model (e.g. J48 Tree, SVM, Logistic Regression, Linear Regression)?
2
1
Scott Tsai's profile photoLevente Varga's profile photo
Add a comment...
 
Hello all, I am working on a predictive modeling project at the moment and I am using #Python  to preprocess the data, which is complicated enough that the script to process it will be several hundred lines of code to get it into a format that could be fed into a machine learning algorithm. 

My original plan was to convert the data to be used with Weka, which is a Java machine learning platform, but it was suggested to me that if I was already this embedded in Python for the project that I might as well step it up and learn to do the machine learning modeling with Python.

So my question is which libraries in Python would provide the best tools to implement, test, and compare several different types of predictive model (e.g. J48 Tree, SVM, Logistic Regression, Linear Regression)?
1
Travis McKee's profile photoErich Schubert's profile photoCameron Craddock's profile photoJohn Douglas Porter's profile photo
7 comments
 
Its not about age. IMHO weka suffers from some design decisions that may be well due to the first java version it was written on. It could be much faster, but you would have to rewrite a lot and break compatibility. So newer may yield some benefits - and scikit learn for sure has a large development community, it will continue to grow fast. 
Add a comment...
 
Hello there folks, I had a question about Python style for-loop style. What would be good style to emulate the nested c-style for loops below in Python? I did it with a couple of while loops and counters for now, but it is very ugly. 

for (int i =0; i < N; i++) { 
       for (int j = i; j < N; j++) { 
              access(list[j])
       } 


Or in pseudo code: 

for each element i in list do: 
      for each element j from i to end of list do: 
1
Travis McKee's profile photoMarc Poulin's profile photoEhud Tamir's profile photoKen Conaway's profile photo
6 comments
 
You should try itertools.combinations. It was designed exactly for that:
import itertools
for x, y in itertools.combinations(my_list, 2):
do_something(x, y) 

This will generate all distinct pairs from the input list. 
Add a comment...

Travis McKee

Shared publicly  - 
 
 

8 bit Star Wars
1
Add a comment...

Travis McKee

Shared publicly  - 
1
Add a comment...
Have him in circles
115 people
Dan Rose's profile photo
Hannah Close's profile photo
Dillon Hanley's profile photo
Jim Reach's profile photo
Alistair McKee's profile photo
Clint Pearson's profile photo

Travis McKee

Shared publicly  - 
 
 
SANITARYUM is using is using its influence for GOOD!
Help us make a difference and help fight muscle disease.  
Donate @ our custom MDA webpage:www2.mda.org/goto/sanitaryum14
or here: http://bit.ly/SanitaryumMDA

C'mon Sanitaryum Patients - WE CAN DO IT!
More pics at ▀▄▀ +Clean Funny Pics & Humor | Sanitaryum.com ▄▀▄


#funny #lol #funnypictures #funnypics #lolz #haha #hilarious #humor #funnypics #smile #comic #FAIL   #MDA  #goodcause #donate  
1
Add a comment...

Travis McKee

Newbie Discussion  - 
 
Ok I feel like I'm losing my mind, since I've done this a hundred times. 
When I redirect this input: 

5 6 7 8 9 10
6
1 2
2 3
3 4
1 5
5 4
4 5
2
4
5

into this program: 

int temp; 
scanf("%d", &temp); 
while (temp) {
        printf("%d\n", temp);
        scanf("%d", &temp); 


I get an infinite loop of: 

5
5
5
5
...
segfault. 

Why does scanf not consume the input? 
1
Florian Philipp's profile photoRobert Dailey's profile photoOndrej Preclik's profile photoJon Dubovsky's profile photo
18 comments
 
These are specific competition problems, and in general people use scanf all the time in competition because you don't want to waste time reconfiguring your code after the fact. 

I personally don't like the problems much, I think they are contrived and tend to instill poor programming practice, but they are required for a class I am in so I don't have much of a choice at this point. 
Add a comment...

Travis McKee

Shared publicly  - 
 
 
"A boy left his bike chained to a tree when he went away to war in 1914. He never returned, leaving the tree no choice but to grow around the bike."
1
Add a comment...

Travis McKee

Shared publicly  - 
 
 
This is a PSA:
2
Paul Gray's profile photo
 
I like hushpuppy shoes, very soft and comfortable.
Add a comment...
People
Have him in circles
115 people
Dan Rose's profile photo
Hannah Close's profile photo
Dillon Hanley's profile photo
Jim Reach's profile photo
Alistair McKee's profile photo
Clint Pearson's profile photo
Work
Skills
C++, Linux, Piano, Guitar, Object Oriented Design
Employment
  • New York University
    Computer Science Tutor, 2013 - present
    Answering questions for introductory computer science students.
  • Garnet River LLC
    Programmer Intern, Systems Integration Division, 2013 - 2013
  • SUNY Adirondack
    Computer Science Tutor, 2012 - 2013
    Tutoring Math and Computer Science
  • GNM Stonemasonry
    Stonemason, 2002 - 2011
    Building fireplaces, patios etc.
  • Centerplate
    Ski Tech, 2005 - 2011
    Tuning up skis (and snowboards), setting people up to ride etc.
Basic Information
Gender
Male
Other names
Jarvis
Story
Tagline
Pretty cool guy.
Education
  • New York University
    Computer Science, 2013 - present
    Basic Algorithms, Operating Systems, Linear Algebra, Algorithmic Problem Solving, Data Mining for Business Analytics
  • SUNY Adirondack
    Computer Science, 2011 - 2013
    Data Structures in C++, Multivariate Calculus, Object Oriented Design, Computer Systems and Design, 8086 Assembly Language, Discrete Mathematics, Economics, Financial Accounting
Links