Inference

Scoring code

Challenge

The challenge will ask participants to measure the ionization fraction for each redshift x(z) and the corresponding  uncertainly P(x|z).

Scoring

The score will be separate for datasets 1 and 2. 

For each redshift, the score will be the submitted posterior P(x|z) evaluated at the true input value, x'. Probabilities corresponding to different redshifts will be multiplied together to give the dataset score. 



During the challenge, no score will be returned to the teams. A format validation service will provide feedback on whether a submission has been received and it can be read correctly by the scoring routine. The last submission for each dataset will be scored.