o }oi@sddlmZddlmZddlZddlZddlZddlZddl m Z ddl m Z ddl mZd d dZd!ddZd"ddZGddde ZdS)#) annotations)KeysViewN) BasePruner)StudyDirection) TrialStatetrial'optuna.trial.FrozenTrial' directionrreturnfloatcCs6tjt|jtd}|tjkrt|St |S)Ndtype) npasarraylistintermediate_valuesvaluesr rMAXIMIZEnanmaxnanmin)rr rrN/home/ubuntu/.local/lib/python3.10/site-packages/optuna/pruners/_percentile.py(_get_best_intermediate_result_over_stepss   rcompleted_trials list['optuna.trial.FrozenTrial']stepint percentile n_min_trialscsdt|dkr tdfdd|D}t||krtjS|tjkr%d|}tttj |td|S)NrzNo trials have been completed.cs g|] }|jvr|jqSr)r).0trrr "szC_get_percentile_intermediate_result_over_trials..dr ) len ValueErrormathnanrrr r nanpercentilearray)rr rrrrrr!r/_get_percentile_intermediate_result_over_trialss     r*intermediate_steps KeysView[int]n_warmup_stepsinterval_stepsboolcs>||||}|dksJtfdd|d}||kS)Nrcs||kr |kr |S|S)Nr)second_last_stepsr!rr>sz,_is_first_in_interval_step..) functoolsreduce)rr+r-r.nearest_lower_pruning_stepr0rr!r_is_first_in_interval_step4s  r7c@s2eZdZdZ   ddddddZdddZdS)PercentilePruneraI Pruner to keep the specified percentile of the trials. Prune if the best intermediate value is in the bottom percentile among trials at the same step. Example: .. testcode:: import numpy as np from sklearn.datasets import load_iris from sklearn.linear_model import SGDClassifier from sklearn.model_selection import train_test_split import optuna X, y = load_iris(return_X_y=True) X_train, X_valid, y_train, y_valid = train_test_split(X, y) classes = np.unique(y) def objective(trial): alpha = trial.suggest_float("alpha", 0.0, 1.0) clf = SGDClassifier(alpha=alpha) n_train_iter = 100 for step in range(n_train_iter): clf.partial_fit(X_train, y_train, classes=classes) intermediate_value = clf.score(X_valid, y_valid) trial.report(intermediate_value, step) if trial.should_prune(): raise optuna.TrialPruned() return clf.score(X_valid, y_valid) study = optuna.create_study( direction="maximize", pruner=optuna.pruners.PercentilePruner( 25.0, n_startup_trials=5, n_warmup_steps=30, interval_steps=10 ), ) study.optimize(objective, n_trials=20) Args: percentile: Percentile which must be between 0 and 100 inclusive (e.g., When given 25.0, top of 25th percentile trials are kept). n_startup_trials: Pruning is disabled until the given number of trials finish in the same study. n_warmup_steps: Pruning is disabled until the trial exceeds the given number of step. Note that this feature assumes that ``step`` starts at zero. interval_steps: Interval in number of steps between the pruning checks, offset by the warmup steps. If no value has been reported at the time of a pruning check, that particular check will be postponed until a value is reported. Value must be at least 1. n_min_trials: Minimum number of reported trial results at a step to judge whether to prune. If the number of reported intermediate values from all trials at the current step is less than ``n_min_trials``, the trial will not be pruned. This can be used to ensure that a minimum number of trials are run to completion without being pruned. r)rrr n_startup_trialsrr-r.rr NonecCsd|kr dksntd|d|dkrtd|d|dkr+td|d|dkr7td |d|dkrCtd |d||_||_||_||_||_dS) Ngr#zCPercentile must be between 0 and 100 inclusive, but got percentile=.rzFNumber of startup trials cannot be negative, but got n_startup_trials=zBNumber of warmup steps cannot be negative, but got n_warmup_steps=r:zBPruning interval steps must be at least 1, but got interval_steps=zFNumber of trials for pruning must be at least 1, but got n_min_trials=)r% _percentile_n_startup_trials_n_warmup_steps_interval_steps _n_min_trials)selfrr;r-r.rrrr__init__s2       zPercentilePruner.__init__study'optuna.study.Study'rrr/c Cs|jdtjfd}t|}|dkrdS||jkrdS|j}|dur#dS|j}||kr,dSt||j ||j s9dS|j }t ||}t |rHdSt||||j|j} t | rYdS|tjkrb|| kS|| kS)NF)deepcopystatesrT) get_trialsrCOMPLETEr$r? last_stepr@r7rkeysrAr rr&isnanr*r>rBrr) rCrErrn_trialsrr-r best_intermediate_resultprrrprunes8     zPercentilePruner.pruneN)r9rr:) rr r;rr-rr.rrrr r<)rErFrrr r/)__name__ __module__ __qualname____doc__rDrQrrrrr8FsD$r8)rrr rr r ) rrr rrrrr rrr r ) rrr+r,r-rr.rr r/) __future__rcollections.abcrr4r&numpyroptunaoptuna.prunersroptuna.study._study_directionroptuna.trial._staterrr*r7r8rrrrs