odow / SDDP.jl

A JuMP extension for Stochastic Dual Dynamic Programming
https://sddp.dev
Other
305 stars 61 forks source link

TTFX #685

Closed odow closed 2 months ago

odow commented 1 year ago
using SnoopCompile
tinf = @snoopi_deep include("docs/src/examples/Hydro_thermal.jl")
import ProfileView
fg = flamegraph(tinf)
ProfileView.view(fg)
itrigs = inference_triggers(tinf)
mtrigs = accumulate_by_source(Method, itrigs)
modtrigs = filtermod(SDDP, mtrigs)

image

odow commented 2 months ago

This continues to get better without us doing anything. It's hard to precompile a call to SDDP.train without embedding an optimizer in our package...

Closing as won't fix.