Spatialize huge graphs in CLI with graphology then explore them on the web with Sigma.js.
git clone https://github.com/medialab/bigma
cd bigma
npm installThe script requires an edges CSV structured as Souce,Target,Weight.
# Run NB_FA2_ITERATIONS iterations with ForceAtlas2 and render PNG snapshots every NB_ITERATIONS_BETWEEN_MINIATURE_PNG_SNAPSHOTS:
node spatialize-network.js <EDGELIST_AS_SOURCE-TARGET-WEIGHT.CSV> <NB_FA2_ITERATIONS> <NB_ITERATIONS_BETWEEN_MINIATURE_PNG_SNAPSHOTS>
# The script can restart new iterations after a full run or a run interrupted by CTRL+C, using the positions CSV file dumped by the previous run (which should be named such as "ORIGINALFILE.CSV_positions_after_N_FA2Iterations.csv"):
node spatialize-network.js <ORIGINALFILE.CSV_positions_after_N_FA2Iterations.csv> <NB_EXTRA_FA2_ITERATIONS> <NB_ITERATIONS_BETWEEN_MINIATURE_PNG_SNAPSHOTS>
# Graphs are plotted by default with an identical size for all nodes, you can add an extra argument to use the nodes degrees as size:
node spatialize-network.js <EDGELIST_AS_SOURCE-TARGET-WEIGHT.CSV> <NB_FA2_ITERATIONS> <NB_ITERATIONS_BETWEEN_MINIATURE_PNG_SNAPSHOTS> 1
If you have ImageMagick's convert tool installed, you can visualize the spatialization evolution as a gif by running:
./build_gif.sh <ORIGINAL_EDGELIST.CSV>
npm startThen visit http://localhost:3000/.
First load in the top-right box the positions CSV produced by the script, then the original edges file.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent