poisson_learning

https://github.com/mihailescum/poisson_learning

Science Score: 44.0%

This score indicates how likely this project is to be science-related based on various indicators:

✓
CITATION.cff file
Found CITATION.cff file
✓
codemeta.json file
Found codemeta.json file
✓
.zenodo.json file
Found .zenodo.json file
○
DOI references
○
Academic publication links
○
Academic email domains
○
Institutional organization owner
○
JOSS paper metadata
○
Scientific vocabulary similarity
Low similarity (10.0%) to scientific vocabulary

Last synced: 6 months ago · JSON representation ·

Repository

Basic Info

Host: GitHub
Owner: mihailescum
License: mit
Language: Python
Default Branch: main
Size: 30.6 MB

Statistics

Stars: 1
Watchers: 1
Forks: 0
Open Issues: 0
Releases: 0

Created about 4 years ago · Last pushed over 3 years ago

Metadata Files

Readme License Citation

About

This repository contains all code related to my master's thesis on the continuum limits of Poisson learning, handed in October 2022 at the University of Bonn.

Requirements

You will need to install the following packages:

python(3.9.7)
numpy (1.12.2)
pandas (1.3.4)
scipy (1.7.2)
graphlearning (1.1.3)
matplotlib (3.4.3)
seaborn (0.11.2)
hdf5 (1.12.1)

How to run

In the examples folder you find several preconfigured experiments that you can run. The results will be written to a results folder, which you should create in advance in your working directory. If you run the corresponding <name>_results.ipynb notebooks afterwards, the generated plots will be saved to a plots folder, which you should also create in advance.

For some of the experiments you can configure the number of threads to use using the NUM_THREADS variable at the begining of the scripts. Moreover, you have the possibility to specify a SEED_RANGE. Each value in this range will be the seed of a separate independent trial of the experiment you run, therefore you can test the experimental results for different, yet deterministic, inputs.

Description of the experiments:

line: 1D experiment for Poisson learning on the unit interval (0, 1) with two labeled nodes, one at 0.4, one at 0.8.
p_line: Same as line, only that we do p-Poisson learning for a range of values of p.
one_circle: 2D experiment for Poisson learning on the unit disc B_1(0) with two labeled nodes, one at (-2/3, 0) and one at (2/3, 0).
p_one_circle: Same as one_circle, only that we do p-Poisson learning for a range of values of p.
real_data: p-Poisson learning on the two real world data sets MNIST and FashionMNIST.

Owner

Login: mihailescum
Kind: user

Repositories: 3
Profile: https://github.com/mihailescum

Citation (CITATION.cff)

cff-version: 1.2.0
message: "If you use this software, please cite it as below."
authors:
  - family-names: Mihailescu
    given-names: Max Eric
title: "Poisson Learning Experiments"
version: 1.0
date-released: 2022-10-05

ecosyste.ms

Data

Tools

Indexes

Applications

Experiments

Open Source Science