On the read me - the link of the vessel power prediction is still to phase 1 - it should be updated with the link from phase 2 which includes both real and synthetic data https://zenodo.org/records/7684813
Also according to the paper "The evaluation sets will not be released, but will be hosted on permanent leaderboards on the Grand-Challenge platform 7. Should the leaderboards close for any reason, the evaluation sets will be similarly released via Zenodo. The reason for keeping the evaluation sets private is to ensure a truly clean ’out-of-domain generalisation
scenario’ and avoid any possible, even unintentional, data leakage."
However, the leader board is permanently closed and evaluation sets have not been released on Zenodo, How are we supposed to evaluate our models?
On the read me - the link of the vessel power prediction is still to phase 1 - it should be updated with the link from phase 2 which includes both real and synthetic data https://zenodo.org/records/7684813
Also according to the paper "The evaluation sets will not be released, but will be hosted on permanent leaderboards on the Grand-Challenge platform 7. Should the leaderboards close for any reason, the evaluation sets will be similarly released via Zenodo. The reason for keeping the evaluation sets private is to ensure a truly clean ’out-of-domain generalisation
scenario’ and avoid any possible, even unintentional, data leakage."
However, the leader board is permanently closed and evaluation sets have not been released on Zenodo, How are we supposed to evaluate our models?