Naive SVM library in Python
Project description
By Andrew Tulloch (http://tullo.ch)
Introduction
This is a basic implementation of a soft-margin kernel SVM solver in Python using numpy and cvxopt.
See http://tullo.ch/articles/svm-py/ for a description of the algorithm used and the general theory behind SVMs.
Demonstration
Run bin/svm-py-demo –help.
∴ bin/svm-py-demo --help usage: svm-py-demo [-h] [--num-samples NUM_SAMPLES] [--num-features NUM_FEATURES] [-g GRID_SIZE] [-f FILENAME] optional arguments: -h, --help show this help message and exit --num-samples NUM_SAMPLES --num-features NUM_FEATURES -g GRID_SIZE, --grid-size GRID_SIZE -f FILENAME, --filename FILENAME
For example,
bin/svm-py-demo --num-samples=100 --num-features=2 --grid-size=500 --filename=svm500.pdf
yields the image
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
svmpy-0.3.tar.gz
(3.9 kB
view hashes)
Built Distribution
svmpy-0.3.macosx-10.8-x86_64.exe
(66.9 kB
view hashes)
Close
Hashes for svmpy-0.3.macosx-10.8-x86_64.exe
Algorithm | Hash digest | |
---|---|---|
SHA256 | 7842b445b449f2108462311afa054690ee056f2e9c050acd815281fa6ce7bc33 |
|
MD5 | 1f2f463d541c6751750975f7d890a583 |
|
BLAKE2b-256 | 2f743d2a8fb7c14193beed97ae54fe8c4b02fe100cf434383fb0aa1b1b04f6b7 |