Synthetic 1D Multi Biased Distribution Generation
A program for generating multi biased distributions. This tool can be used for quickly generating biased random values for training data in machine learning. This is especially useful in early stages of neural network development or when you quickly want to try out a certain function without a sufficient dataset.
Methods used by the program:
-
Integer Partitioning "In number theory and combinatorics, a partition of a positive integer n, also called an integer partition, is a way of writing n as a sum of positive integers." (https://en.wikipedia.org/wiki/Partition_(number_theory))
-
Kernel Density Estimation (KDE) "A non-parametric way to estimate the probability density function of a random variable." (https://en.wikipedia.org/wiki/Kernel_density_estimation)
Clone the project:
git clone https://github.com/Irreq/multi-bias-gen.git
Move to the project folder:
cd multi-bias-gen
Install dependencies:
pip3 install requirements.txt
See:
python3 demo.py