Adeline Hillier comments

Results 7 comments of


                                            Adeline Hillier

Update Oceananigans compat

I'm baffled by the compilation errors here.

Large forward maps make Julia crash

I just realized that `EnsembleKalmanProcesses.jl` can't take non-`Matrix` types for the covariance matrix, so I'm also leaning towards 3.

Doing all training/work with buoyancy instead of temperature?

Sounds fine to me! Though can't we easily compute the buoyancy flux from the temp flux? Also, buoyancy values are a bit less interpretable and the OceanTurb.jl simulations are set...

Use `.dataset` rather than `.field_time_serieses`?

That sounds like a good idea. "field_time_serieses" is a lot to type.

User interface to `Scented` and `Unscented` Kalman inversions

Yes, good ideas! I think the existing `UnscentedKalmanInversionPostprocess` can be absorbed into `IterationSummary`. The arrays returned by `UnscentedKalmanInversionPostprocess` are ```Julia `mean :: Matrix{Float64}`: `N_iter` × `N_θ` mean matrix `cov ::...

Add regridding example to `intro_to_observations.jl`

Brilliant. Just brilliant.

Proper `BatchedObservations` with optional weights?

Then in the scripts "`observations`" could refer to a single instance of `SyntheticObservations`, a vector of `SyntheticObservations` or an instance of `BatchedObservations`. Do you think this would create confusion?