Awesome

PyNeRV

A Python wrapper of the dimensionality reduction algorithm Neighbor Retrieval Visualizer (NeRV)

Compile

Set up the paths in Makefile then make.

Usage example

import numpy as np
import matplotlib.pyplot as plt
from nerv import NeRV

data = np.genfromtxt('dredviz-1.0.2/spheredata.dat', skip_header=1)
color = np.genfromtxt('dredviz-1.0.2/spheredata.col', skip_header=1)
nerv_obj = NeRV(random_state=0)
nerv_obj.fit(data)
print('Final cost = %g' % nerv_obj.cost_)
plt.scatter(*nerv_obj.embedding_.T, c=color)
plt.show()

which shows (right)

<img src="https://raw.githubusercontent.com/ziyuang/pynerv/master/spheredata.png" height="336px"></img>  <img src="https://raw.githubusercontent.com/ziyuang/pynerv/master/spheredata_nerv.png" height="336px"></img>

Reference

@article{venna2010information,
  title={Information retrieval perspective to nonlinear dimensionality reduction for data visualization},
  author={Venna, Jarkko and Peltonen, Jaakko and Nybo, Kristian and Aidos, Helena and Kaski, Samuel},
  journal={Journal of Machine Learning Research},
  volume={11},
  number={Feb},
  pages={451--490},
  year={2010}
}