Visualization framework

Visualize component

Command line parameters of the framework. TODO

Parameter	Description	Example
n_samples	Number of samples for each target class. If not provided, all samples will be generated.	--n_samples 10
parameters	Visualization specific parameters in json format.	--parameters={ \"colorMap\": \"OrangeBlue\", \"bottomMargin\": 25, \"scaleCoeff\": 100, \"correlationOrdering\": true}
parametersFile	File with visualization specific parameters in json format.	--parametersFile config/radial_default.json

Breast cancer

Source: https://archive.ics.uci.edu/ml/datasets/breast+cancer+wisconsin+(diagnostic)

./visualize.py breastCancer blobs _output/breastCancer/blobs --quantize

./visualize.py breastCancer radial _output/breastCancer/radial --n_samples 100 --quantize --parametersFile config/radial_default.json

./visualize.py breastCancer faces _output/breastCancer/faces --groups med --n_samples 100 --quantize

Nutrients

Source: https://github.com/syntagmatic/parallel-coordinates/blob/master/examples/data/nutrients.csv

./visualize.py nutrients blobs _output/nutrients/blobs

./visualize.py nutrients radial _output/nutrients/radial --n_samples 10 --quantize --parametersFile config/radial_default.json

./visualize.py nutrients faces _output/nutrients/faces --groups med --n_samples 100 --quantize

Recipes

Source: https://osf.io/cpfwj

Source: https://www.kaggle.com/datasets/elisaxxygao/foodrecsysv1?select=core-data_recipe.csv

./visualize.py recipes blobs _output/recipes/blobs

./visualize.py recipes radial _output/recipes/radial --n_samples 100 "--parameters={ \"colorMapPath\": \"data/ingredientsMap.csv\"}"

Spotify

Source: https://www.kaggle.com/datasets/zaheenhamidani/ultimate-spotify-tracks-db

Example:

./visualize.py spotify blobs _output/spotify/blobs --normalize

./visualize.py spotify radial _output/spotify/radial --quantize --n_samples 100 --parametersFile config/radial_default.json

./visualize.py spotify faces _output/spotify/faces --groups med --n_samples 100 --normalize

Takes spotify dataset (with normalization) and visualize it like blobs. Images are saved into "output" folder.

Average component

Examples:

./average.py _output/breastCancer/radial
./average.py _output/recipes/radial
./average.py _output/spotify/blobs
./average.py _output/spotify/radial

Selected averaging for datasets

./visualize.py breastCancer radial _output/breastCancer/radial --n_samples 100 --quantize --groups med --parametersFile config/radial_default.json

python visualize.py nutrients radial _output/nutrients/radial --quantize --n_samples 100 --groups med --parametersFile config/radial_default.json

Takes folders in "output" folder and makes average of all image in this subfolder.

Authors

• David Bernhauer

• David Hoksza

• Ladislav Peška

• Ivana Sixtová

• Tomáš Skopal