Skip to content

Learning from Networks@UniPD laboratory project. Analysis of the correlation between graph domain and motif distribution

Notifications You must be signed in to change notification settings

gioffo/lfn_project

Repository files navigation

Learning from Networks Project

Project structure

  • /data/ -> Folder for the datasets, before(.csv, .tar.gz, etc..) and after (.graphml) preprocessing
  • /exp_shuffles/ -> Folder for PDA experiment result binaries
  • /experiment/ -> Folder for main experiment result binaries
  • /figures/ -> Folder for saved figures of plots
  • /test_notebooks/ -> Folder for the notebooks used for testing
  • preprocessing.ipynb -> Preprocessing of the datasets
  • run_experiment.ipynb -> Compute counts and significance of motifs in the datasets and save the results as binary files.
  • compare_z_scores.ipynb -> One on one comparison of motifs
  • read_and_plot.ipynb -> Batch processing and plotting
  • example.ipynb -> Example experiment on Escherichia Coli dataset
  • performance_degradation_analysis.ipynb -> Experiment on performance degration of motif_significance() when using speed-up parameters

Installation and run

Hopefully this works:

conda create --name fl-gt --file requirements.txt
conda activate fl-gt
jupyter notebook

About

Learning from Networks@UniPD laboratory project. Analysis of the correlation between graph domain and motif distribution

Topics

Resources

Stars

Watchers

Forks