Feature Extraction with KNN

Python implementation of feature extraction with KNN.

And @momijiame updated my implementation. I recommend to use this:
https://github.com/momijiame/gokinjo

pip install gokinjo

The following is R implementation:
http://davpinto.com/fastknn/articles/knn-extraction.html#understanding-the-knn-features

• Python 3.x
  • numpy
  • scikit-learn
  • scipy

git clone [email protected]:upura/knnFeat.git
cd knnFeat
pip install -r requirements.txt

Notebook version can be seen here.

import numpy as np
%matplotlib inline
import matplotlib.pyplot as plt

x0 = np.random.rand(500) - 0.5
x1 = np.random.rand(500) - 0.5
X = np.array(list(zip(x0, x1)))
y = np.array([1 if i0 * i1 > 0 else 0 for (i0, i1)  in list(zip(x0, x1))])

from knnFeat import knnExtract
newX = knnExtract(X, y, k = 1, folds = 5)

Quote from here.

It generates k * c new features, where c is the number of class labels. The new features are computed from the distances between the observations and their k nearest neighbors inside each class, as follows:

1. The first test feature contains the distances between each test instance and its nearest neighbor inside the first class.
2. The second test feature contains the sums of distances between each test instance and its 2 nearest neighbors inside the first class.
3. The third test feature contains the sums of distances between each test instance and its 3 nearest neighbors inside the first class.
4. And so on.

This procedure repeats for each class label, generating k * c new features. Then, the new training features are generated using a n-fold CV approach, in order to avoid overfitting.

flake8 .

pytest

pytest -v -m 'success' --cov=.

MIT

upura