Package for the data-driven representation of non-linear dynamics over manifolds based on a statistical distribution of local phase portrait features. Includes specific example on dynamical systems, synthetic- and real neural datasets. https://agosztolai.github.io/MARBLE/
The resampling option in further points sampling did not reuse points. This lead to sparser graphs. This is now fixed and does not affect the results of the paper.
After further points sampling, the sample_ind attribute is now sorted.
plotting.embedding() now allows plotting trajectories and color gradients to signify the arrow of time. To use this, one needs to add a 'labels' field in construct_dataset() containing the time indices of the data points.
The resampling option in further points sampling did not reuse points. This lead to sparser graphs. This is now fixed and does not affect the results of the paper.
After further points sampling, the sample_ind attribute is now sorted.
plotting.embedding() now allows plotting trajectories and color gradients to signify the arrow of time. To use this, one needs to add a 'labels' field in construct_dataset() containing the time indices of the data points.