more nitpicks on end-to-end

automl / TabPFN

Official implementation of the TabPFN paper (https://arxiv.org/abs/2207.01848) and the tabpfn package.

http://priorlabs.ai

Apache License 2.0

1.22k stars 109 forks source link

more nitpicks on end-to-end #28

Closed amueller closed 1 year ago

amueller commented 1 year ago

Another nitpick on the end-to-end training and evaluation. Currently the TabPFNClassifier is hard-coded to read from f'models_diff/prior_diff_real_results{add_name}_n_{i}_epoch_{e}_{eval_addition}.pkl' but I don't see any code that would write a model to that path.

amueller commented 1 year ago

ok never mind that wasn't the problem.