score_theme | R Documentation |
A helper utility to score a collection of forecasts efficiently for challenges hosting submitted forecasts on S3 buckets. Scores are automatically streamed to an "scores" bucket in parquet format. A provenance bucket is used to allow function to skip forecast-target combinations that have already been scored.
score_theme(
theme,
s3_forecasts,
s3_targets,
s3_scores,
s3_prov,
local_prov = paste0(theme, "-scoring-prov.csv"),
endpoint = "data.ecoforecast.org",
s3_inv = arrow::s3_bucket("neon4cast-inventory", endpoint_override = endpoint)
)
theme |
which theme should be scored? |
s3_forecasts |
a connection from arrow::s3_bucket |
s3_targets |
a connection from arrow::s3_bucket |
s3_scores |
a connection from arrow::s3_bucket where scores will be written. This connection requires write access, e.g. by specifying |
s3_prov |
a connection from arrow::s3_bucket |
local_prov |
path to local csv file which will be used to store provenance until theme is finished scoring. |
endpoint |
endpoint must be passed explicitly for s3_forecast bucket |
s3_inv |
parquet-based S3 inventory of forecast filepaths |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.