-
-
Save WanyingF/33b83cbd418f36f4d263ecbdc54f7ccd to your computer and use it in GitHub Desktop.
optimise hyperparameters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
from xgboost import XGBClassifier | |
from platypus import NSGAII, Problem, Real, Integer | |
import numpy as np | |
from sklearn.metrics import accuracy_score, roc_auc_score | |
from sklearn.model_selection import train_test_split | |
import sys | |
X, y = np.arange(10).reshape((5, 2)), range(5) | |
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42) | |
def ml(**kwargs) -> [float, float]: | |
learner = XGBClassifier(**kwargs) | |
try: | |
learner.fit(X_train, y_train) | |
y_predict = learner.predict(X_test) | |
y_proba = learner.predict_proba(X_test) | |
y_pred_prob = np.array(y_proba) | |
if len(set(y_test)) == 2 and len(y_pred_prob.shape) > 1: | |
y_pred_prob = y_pred_prob[:, 1] | |
acc = accuracy_score(y_test, y_predict) | |
roc = roc_auc_score(y_test, y_pred_prob, multi_class='ovr' | |
return [acc, roc] | |
except Exception as e: | |
print("error! ", e) | |
return [sys.float_info.min, sys.float_info.min] | |
class HyperTuneProblem(Problem): | |
def __init__(self): | |
super(HyperTuneProblem, self).__init__(3, 2) | |
self.types[:] = [Integer(10, 100), Integer(0, 100), Integer(1, 100)] | |
def evaluate(self, solution): | |
param_values = solution.variables[:] | |
param_names = ["n_estimators", "max_delta_step", "num_parallel_tree"] | |
params = dict(zip(param_names, param_values)) | |
solution.objectives[:] = ml(**params) | |
# define the range of each parameter, here we are listing the example range for n_estimator, max_delta_step and num_parallel_tree | |
platypus_parameters = [Integer(10, 100), Integer(0, 100), Integer(1, 100)] | |
problem = HyperTuneProblem() | |
problem.types[:] = platypus_parameters | |
algorithm = NSGAII(problem, population_size=10) | |
algorithm.run(10) | |
for solution in algorithm.result: | |
print(solution.objectives) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment