Python serverside toolkit
Project description
Weighted Centroid Distance
A new model assessment technique for multi-class classification tasks in machine learning.
For the statistical theory, please see the white paper in /paper
. It is auto
generated from .tex
to .pdf
.
How to use
# Python
from weighted_centroid_distance import WeightedCentroidDistance
distribution = WeightedCentroidDistance.get_distribution([1, 2, 3, 3, 4, 5, 5, 5], inverse=False)
wcd = WeightedCentroidDistance(distribution=distribution)
y____ = [1, 1, 2, 2, 3, 3, 3, 4, 4, 5, 5, 5, 5, 5, 5]
y_hat = [1, 1, 2, 2, 3, 3, 4, 4, 4, 4, 5, 1, 5, 5, 2]
res = wcd.get_distance(y=y____, y_hat=y_hat)
print("res: ", res)
# 0.12086457016125143
Todo
- Create graphs for all parts of the algorithm defined in the paper
- Improve python script, and make a versions for other languages.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Close
Hashes for weighted_centroid_distance-1.0.0.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 44cad706f1031997377cf44eefac33c5ad3e6adddc63df01aeba5d7bf6fe9298 |
|
MD5 | 65ed554326bd48cb8c512ba29c6ce97a |
|
BLAKE2b-256 | 26d77a373e19d516c83f880841295f2849a745488a6eb4946def6d164ff705d4 |
Close
Hashes for weighted_centroid_distance-1.0.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 4533f586d586ee55ecc39e592ee3e7996b362fead975a300bb57554d241a64ed |
|
MD5 | 71ce03e4a4644ec0d24cd092a81522b9 |
|
BLAKE2b-256 | 4d1de3a61cea34d04fa2513a528305fff32147d0b37dd8bbd20be5587a31eacb |