o /wix@sHddlmZddlZddlZdd d ZdddZdddZdddZdS)) annotationsNrank_i_loss_vals np.ndarrayrank_i_indices subset_sizeintreference_pointreturncCs:|jddkr||jdksJ|jd}t|jd}|}tj|tjddf|dd}tj|td}t|D]_} tj ||dd} t | } ||| || <|| } tj || t d} d| | <|| }|| }|| }t | d|d| df|d| df<t | d|| ddf|| ddf<q;|S)NraxisdtypeF)shapenparangecopyrepeatnewaxiszerosrrangeprodargmaxonesboolminimum)rrrrn_trialssorted_indicessorted_loss_vals rect_diagsselected_indicesicontribs max_index loss_valskeepr(U/home/ubuntu/sommelier/.venv/lib/python3.10/site-packages/optuna/_hypervolume/hssp.py_solve_hssp_2ds&     (*r*r$pareto_loss_values selected_vecsc Cstjj|dd|dd}d}t| }|D]/}|||kr q|||d<tjj||dd}t|s:||ntj||<t|||}q|S)aLazy update the hypervolume contributions. S=selected_indices - {indices[max_index]}, T=selected_indices, and S' is a subset of S. As we would like to know argmax H(T v {i}) in the next iteration, we can skip HV calculations for j if H(T v {i}) - H(T) > H(S' v {j}) - H(S') >= H(T v {j}) - H(T). We used the submodularity for the inequality above. As the upper bound of contribs[i] is H(S' v {j}) - H(S'), we start to update from i with a higher upper bound so that we can skip more HV calculations. Nr T) assume_paretog) optuna _hypervolumecompute_hypervolumerargsortrisinfinfmax) r$r+r,r hv_selected max_contrib%index_from_larger_upper_bound_contribr#hv_plusr(r(r)_lazy_contribs_update+s  r9cCs8t|s |d|S|j|kr|S|jddkr"t||||S||jks)J||}|j\}}tj|dd}tj|td}t ||f} t |} t |D]G} tt |} | | || <||  | | <tj|jtd} d| | <|| }| | } || }| |dkr||St||| d| d|}qP||S)Nr r r rFr)risfiniteallsizerr*rrremptyrrrrrrr9)rrrrdiff_of_loss_vals_and_ref_point n_solutions n_objectivesr$r"r,indiceskr%r'r(r(r)_solve_hssp_on_unique_loss_valsPs:       rCc Cs||jkr|Stj|ddd\}}|j}||kr;tj|jtd}d||<t|j|}d||d||<||St||||} || S)ahSolve a hypervolume subset selection problem (HSSP) via a greedy algorithm. This method is a 1-1/e approximation algorithm to solve HSSP. For further information about algorithms to solve HSSP, please refer to the following paper: - `Greedy Hypervolume Subset Selection in Low Dimensions `__ Tr) return_indexr rN)r<runiquerrrrC) rrrrrank_i_unique_loss_valsindices_of_unique_loss_valsn_uniquechosenduplicated_indices$selected_indices_of_unique_loss_valsr(r(r) _solve_hsspys  rL) rrrrrrrrr r) r$rr+rr,rrrr r) __future__rnumpyrr.r*r9rCrLr(r(r(r)s   # %)