It seems like we cannot fit ndarray smaller than shape(-1,3).
e.g.
if we want to fit in some array of shape(-1,2):
And it will call a function in gen_meta_features, and do PCA, but the param is a fixed "3".
Also, is it better that "select_model" returns a list of models with its hyperparams. For now, I need to print out the best N selected models and new these models by my own. If the top models is "LODA (5, 100)", I can't simply do "clf = LODA(5,100)", instead, I need to do "LODA(n_bins=5, n_random_cuts=100)" or "LODA(0.1, 5, 100)"
It seems like we cannot fit ndarray smaller than shape(-1,3). e.g. if we want to fit in some array of shape(-1,2):
And it will call a function in gen_meta_features, and do PCA, but the param is a fixed "3".
Also, is it better that "select_model" returns a list of models with its hyperparams. For now, I need to print out the best N selected models and new these models by my own. If the top models is "LODA (5, 100)", I can't simply do "clf = LODA(5,100)", instead, I need to do "LODA(n_bins=5, n_random_cuts=100)" or "LODA(0.1, 5, 100)"