You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
To support extremely large simulations with dense station grids we should introduce chunking using dask. The fastest way to do this would be:
Chunk the hf and lf outputs. Chunking the hf outputs will be done automatically if Dask chunking in hf-sim #108 is closed, but the lf outputs will get chunked manually.
Use xarray's map_blocks api, which allows us map over chunks in serial. Unlike hf sim, we would want to process these in serial because there is already a parallel layer inside the cb2014_to_fas_amplification_factors code using numba so we don't want to doubly parallelise.
To support extremely large simulations with dense station grids we should introduce chunking using dask. The fastest way to do this would be:
cb2014_to_fas_amplification_factorscode using numba so we don't want to doubly parallelise.