Skip to content

sigsep/sigsep-mus-2018-analysis

Repository files navigation

SISEC MUS 2018 Analysis and Visualization

Installation

install the python3.6 requirements using pipenv: pipenv install

Usage

Aggregate Data

To aggregate the data and convert them to various formats we use a pandas data frame.

python aggregate.py ./EST --out sisec18.pd generates a pandas data frame with all results from the ./EST folder.

If you want to compare several estimates, you can just specify them as multiple command line arguments:

python aggregate.py ./EST1 ./EST2 --out sisec18.pd

or use a parent folder and bash wildcards

python aggregate.py ./ESTIMATES/* --out sisec18.pd

Plots and Statistical Analysis

We provide a jupyter notebook that includes all the results that were used to create the SiSEC evaluation. Also you can run the notebook on google colab.

SiSEC 18 Paper Plots

t.b.a

SiSEC 18 Paper

t.b.a