Meta-datasets generated by the paper "Exploring One Million Machine Learning Pipelines: A Benchmarking Study"

READMEMachine learning pipelines run saved.Columns explanation:<code>seed_i</code>: seed used for the experiments<code>config_id</code>: id of the configuration<code>fold</code>: fold of the dataset<code>config_hash</code>: hash value for the configurations<code>duration</code>: durations of the run<code>start_time</code>: start time<code>end_time</code>: end time<code>dataset</code>: the name of the dataset<code>status</code>: Status of the run. If it succeeds or not.<code>[metric_name]_[set split]</code>: performance of the trained model on the set (train, val, test). For example, "f1_weighted_test" is the F1 Score of the trained pipeline on the test set.The remaining columns contain configuration space similar to AutoSklean.