Aaron Spring

**Is your feature request related to a problem? Please describe.** currently `lead`s are connected as a timeseries in `HindcastEnsemble`. would like to plot `valid_time` as timeseries not `lead` as current...

implement `set_options(seasonality=None)`

1

**Is your feature request related to a problem? Please describe.** when not using groupby in bias reduction **Describe the solution you'd like** groupby(None) does nothing

implement `train_test_split='fair-sliding'/'fair-all'`

1

**Is your feature request related to a problem? Please describe.** New paper Risbey et al. 2021 demonstrates how forecaster choices for time selection in bias reduction impacts skill. Reference: https://t.co/HE4PoROb8Z?amp=1...

aggregate over lead

2

**Is your feature request related to a problem? Please describe.** I would like to verify total precipitation, i.e. precip pr aggregated over lead. I can easily create aggregated forecasts, but...

speedup proposal: no lead loop in HindcastEnsemble but lead and init in obs

12

- create init and lead dimensions for `observation` seamlessly with `cfsv2_ds["o"] = (( "L", "S"), o.sel(T=T))` - this approach is thinking more in `real_time` or `valid_time`, however, we do not...

toy model

3

`s2dverification`, similar to `climpred` but written in R, has a toy model to test expected behaviour of skill (dependence of number of inits, members) on synthetic data. nice-to-have ### References...

easy way to download the data with `climetlab`

6

FYI download and open in python: https://climetlab.readthedocs.io/en/latest/examples/09-weatherbench.html

refactor mask `rps`

2

# Description - `xs.rps` sometimes throws a warning when masking all NaN inputs. often when inputs dont have same variables. Closes #(issue) ## Type of change - [x] refactoring #...

Write xs JOSS paper

17

very easy to create a `CITATION.cff` if the software is on zenodo (which `xs` is not yet): https://github.com/citation-file-format/doi2cff xskillscore is actually used in a few studies already: - all climpred...

multi-dimensional category_edges for contingency, reliability, etc``

5

so far with `xhistogram`, we can only create 1d `category_edges`. however, sometimes we would want to make `category_edges` multi-dimensional (dependent on lon and lat) when e.g. calculating wrt. climatological terciles....