https://projectforum.tudelft.nl/course_editions/60/generic_projects/2545
define the following main parameters in experiment_step.jl
- search_alg_name = "astar"
- options: bfs, vlns, mh
- dataset_ids = [1499, 1510, 1504, 1478]
- ids of the dataset you want to evaluate
- n_runs = 10
- the number of times the experiment should run
- max_pipelines = 100
- the max number of pipelines that should be evaluated
- train_on_n_samples = 300
- the number of samples the pipelines should be trained on
- values = [3]
- the maximum depth of pipelines that should be evaluated
- You can try multiple depths and it will run the experiment for each different depth
run experiment_step.jl, results will be in the db_output folder