Editing
Machine Learning/SVM
(section)
Jump to navigation
Jump to search
Warning:
You are not logged in. Your IP address will be publicly visible if you make any edits. If you
log in
or
create an account
, your edits will be attributed to your username, along with other benefits.
Anti-spam check. Do
not
fill this in!
== Converting the Data == As with most (if not all) data problems, choosing and formatting the data is the most time-consuming step but also one of the most important. One approach for reducing the data is to take a subset; you can use Thomas' perl script to take a sample of some number of the training set and test set, by choosing a random subset of the students and only including lines which include them. You can use the perl script [[Machine_Learning/kdd_sample | sample_training.pl]] to do this, by running: perl sample_training.pl -numitems 100 ~/kdd/algebra_2008_2009_train.txt (assuming your data is located in ~/kdd) For SVM, ultimately we need to format the data in two files: a training file and a test file. Each of these will have a numeric class and several numeric predictors. The general format is as follows: <class> 1:<value> 2:<value> 3:<value> ... with an entry (1:, 2:, 3:,...) for each numeric predictor. For example, 0 1:0 2:0 3:0 4:0 5:0 6:1 7:0 8:0 9:0 10:0 11:0 12:0 13:0 14:0 15:0 16:0 17:0 18:0 Thomas created a [[Machine Learning/convert_features.pl | perl script]] to take a training set and convert it (and the corresponding test set) into the correct format by using "correct on first attempt" as the output class and converting student and problem id into a series of binary flag variables (one for each student and problem, indicating whether this class regards this student or this problem). However, this results in a fairly obscene number of predictor variables, even on a stripped-down dataset. So there is almost certainly a better way. But if you don't have one, you can download this script and run perl convert_features.pl ~/kdd/algebra_2008_2009_train.txt_sample_100_random_students.csv Assuming your data files are in ~/kdd, this will generate output files ~/kdd/algebra_2008_2009_train.txt_sample_10_random_students.csv_converted.txt and ~/kdd/algebra_2008_2009_train.txt_sample_10_random_students.csv_converted.t in the appropriate format.
Summary:
Please note that all contributions to Noisebridge are considered to be released under the Creative Commons Attribution-NonCommercial-ShareAlike (see
Noisebridge:Copyrights
for details). If you do not want your writing to be edited mercilessly and redistributed at will, then do not submit it here.
You are also promising us that you wrote this yourself, or copied it from a public domain or similar free resource.
Do not submit copyrighted work without permission!
To protect the wiki against automated edit spam, we kindly ask you to solve the following CAPTCHA:
Cancel
Editing help
(opens in new window)
Navigation menu
Personal tools
Not logged in
Talk
Contributions
Log in
Request account
Namespaces
Page
Discussion
English
Views
Read
Edit
View history
More
Search
Dig in!
Noisebridge
- Status: MOVED
- Donate
- ABOUT
- Accessibility
- Vision
- Blog
Manual
MANUAL
Visitors
Participation
Community Standards
Channels
Operations
Events
EVENTS
Guilds
GUILDS
- Meta
- Electronics
- Fabrication
- Games
- Music
- Library
- Neuro
- Philosophy
- Funding
- Art
- Crypto
- Documentation/Wiki
Wiki
Recent Changes
Random Page
Help
Categories
(Edit)
Tools
What links here
Related changes
Special pages
Page information