weights_over_time_method
Back to L4.5 | Browse all axes | Browse all options
Axis
weights_over_time_methodon sub-layerL4_5_E_ensemble_diagnostics(layerl4_5).
Sub-layer
L4_5_E_ensemble_diagnostics
Axis metadata
Default:
'stacked_area'Sweepable: False
Status: operational
Operational status summary
Operational: 3 option(s)
Future: 0 option(s)
Options
heatmap – operational
Heatmap of weights (member × time).
L4.5.E weights-over-time rendering heatmap.
This option configures the weights_over_time_method axis on the L4_5_E_ensemble_diagnostics sub-layer of L4.5; output is emitted under manifest.diagnostics/l4_5/L4_5_E_ensemble_diagnostics/ alongside the other selected views.
When to use
Many-member ensembles (> 20) where line / area plots become unreadable.
References
macroforecast design Part 4: ‘diagnostic layers default-off; non-blocking; produce JSON + matplotlib views attached to manifest.diagnostics/.’
Related options: line_plot, stacked_area
Last reviewed 2026-05-05 by macroforecast author.
line_plot – operational
Line plot of weights per member over time.
L4.5.E weights-over-time rendering line_plot.
This option configures the weights_over_time_method axis on the L4_5_E_ensemble_diagnostics sub-layer of L4.5; output is emitted under manifest.diagnostics/l4_5/L4_5_E_ensemble_diagnostics/ alongside the other selected views.
When to use
Default reporting view; readable up to ~10 ensemble members.
References
macroforecast design Part 4: ‘diagnostic layers default-off; non-blocking; produce JSON + matplotlib views attached to manifest.diagnostics/.’
Related options: stacked_area, heatmap
Last reviewed 2026-05-05 by macroforecast author.
stacked_area – operational
Stacked-area plot summing to 1.
L4.5.E weights-over-time rendering stacked_area.
This option configures the weights_over_time_method axis on the L4_5_E_ensemble_diagnostics sub-layer of L4.5; output is emitted under manifest.diagnostics/l4_5/L4_5_E_ensemble_diagnostics/ alongside the other selected views.
When to use
Emphasising member share; ideal for showcasing weight redistribution events.
References
macroforecast design Part 4: ‘diagnostic layers default-off; non-blocking; produce JSON + matplotlib views attached to manifest.diagnostics/.’
Related options: line_plot, heatmap
Last reviewed 2026-05-05 by macroforecast author.