path-trajectory-contract.md
In Dynet: Tidy Temporal Network Analysis

a branching forward family is visually stable

Code
  str(.plot_fingerprint(plot_path_trajectories(.school_paths(), measure = "frequency")))
Output
  List of 9
   $ edge_vertices: int 840
   $ edge_groups  : int 21
   $ edge_widths  : num [1:5] 0.3 0.667 1.033 1.4 2.5
   $ node_x       : num [1:22] 0 1 1 1 1 1 2 2 2 2 ...
   $ node_y       : num [1:22] 0.358 0 0.167 0.333 0.5 ...
   $ node_sizes   : num [1:22] 14 6.36 6.36 7.33 8.08 ...
   $ node_fills   : chr [1:22] "#E69F00" "#F0E442" "#E69F00" "#E69F00" ...
   $ node_labels  : chr [1:22] "Ana (n=19)" "Mira (n=2)" "Jonas (n=2)" "Jonas (n=3)" ...
   $ root_points  : int 0

a backward family is visually stable

Code
  str(.plot_fingerprint(plot_path_trajectories(paths, measure = "time",
    orientation = "vertical")))
Output
  List of 9
   $ edge_vertices: int 600
   $ edge_groups  : int 15
   $ edge_widths  : num [1:8] 0.3 0.483 0.667 0.85 1.4 ...
   $ node_x       : num [1:16] 0 0.167 0.25 0.25 0.25 ...
   $ node_y       : num [1:16] -5 -7 -6 -5 -4 -7 -3 -2 -5 -1 ...
   $ node_sizes   : num [1:16] 4 4 7.78 8.63 10.55 ...
   $ node_fills   : chr [1:16] "#F3AC7E" "#EB9159" "#E58344" "#D66109" ...
   $ node_labels  : chr [1:16] "Kira (t=11.9)" "Leo (t=12.62)" "Cara (t=13.02)" "Finn (t=13.92)" ...
   $ root_points  : int 0