Closed bjack205 closed 2 years ago
Merging #81 (e29a2b8) into main (23c2988) will decrease coverage by
9.78%
. The diff coverage is18.42%
.
@@ Coverage Diff @@
## main #81 +/- ##
==========================================
- Coverage 64.05% 54.27% -9.79%
==========================================
Files 16 16
Lines 1736 1837 +101
==========================================
- Hits 1112 997 -115
- Misses 624 840 +216
Impacted Files | Coverage Δ | |
---|---|---|
src/edmd/edmd.jl | 100.00% <ø> (ø) |
|
src/utils.jl | 68.00% <ø> (-12.96%) |
:arrow_down: |
src/edmd/regression.jl | 19.10% <15.00%> (-26.93%) |
:arrow_down: |
src/edmd/eigenfunctions.jl | 55.07% <42.85%> (ø) |
|
src/moi.jl | 0.00% <0.00%> (-99.15%) |
:arrow_down: |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update b18ed18...e29a2b8. Read the comment docs.
Closes #80
Adds an example of using eDMD to learn the error dynamics for the cartpole, where the actual model has some damping and the nominal one doesn't. Seems to work really well.
This example just shows that the dynamics match closely after simulating the true and learned models for a few seconds, where it's given a trajectory that swings back and forth about the downward equilibrium, without any control.