sklearn.cluster.KMeans中n_init参数是什么意思

人工智能领域优质创作者

博客专家认证

2014-12-20 09:34:38

n_init解释为：
Number of time the k-means algorithm will be run with different centroid seeds. The
final results will be the best output of n_init consecutive runs in terms of inertia.
设置选择质心种子次数，默认为10次。返回质心最好的一次结果（好是指计算时长短）

这个到底是嘛意思？

...全文

1820 1 打赏收藏转发到动态举报

写回复

1 条回复

切换为时间正序

请发表友善的回复…

发表回复

panghuhu250 2014-12-20

打赏
举报

回复 1

每一次算法运行时开始的centroid seeds是随机生成的, 这样得到的结果也可能有好有坏. 所以要运行算法n_init次, 取其中最好的.

AI实战-德国信贷风险评估数据分析预测实例（含18个源代码+48.52 KB完整的数据集）代码手工整理，无语法错误，可运行。包括：18个代码，共225.97 KB；数据大小：1个文件共48.52 KB。使用到的模块： matplotlib.pyplot numpy os pandas seaborn sklearn.model_selection.train_test_split sklearn.model_selection.KFold sklearn.model_selection.cross_val_score sklearn.model_selection.GridSearchCV sklearn.ensemble.RandomForestClassifier sklearn.linear_model.LogisticRegression sklearn.tree.DecisionTreeClassifier sklearn.naive_bayes.GaussianNB sklearn.neighbors.KNeighborsClassifier sklearn.svm.SVC xgboost.XGBClassifier sklearn.metrics.confusion_matrix sklearn.metrics.accuracy_score sklearn.metrics.classification_report category_encoders.OneHotEncoder sklearn.preprocessing.MinMaxScaler sklearn.manifold.TSNE sklearn.decomposition.PCA umap.UMAP tqdm.tqdm sklearn.cluster.KMeans sklearn.metrics.silhouette_score scipy.cluster.hierarchy.dendrogram scipy.cluster.hierarchy.linkage sklearn.cluster.AgglomerativeClustering sklearn.cluster.DBSCAN warnings ucimlrepo.fetch_ucirepo imblearn.over_sampling.SMOTE torch torch.nn torch.utils.data.DataLoader torch.utils.data.TensorDataset sklearn.preprocessing.LabelEncoder sklearn.preprocessing.StandardScaler sklearn.metrics.ConfusionMatrixDisplay sklearn.preprocessing sklearn.metrics.mutual_info_score xgboost xgboost.cv sklearn.model_selection.RepeatedStratifiedKFold plotly.offline.plot plotly.offline plotly.graph_objs plotly.tools plotly.figure_factory sklearn.discriminant_analysis.LinearDiscriminantAnalysis sklearn.metrics.roc_curve sklearn.pipeline.Pipeline sklearn.pipeline.FeatureUnion sklearn.feature_selection.SelectKBest umap plotly.graph_objects snaplib.snaplib.Snaplib google.colab.drive time PIL.Image plotly.offline.init_notebook_mode sklearn.metrics.auc sklearn.metrics.fbeta_score sklearn.ensemble.GradientBoostingClassifier numpy.array keras.datasets.mnist numpy.random.MT19937 numpy.random.RandomState numpy.random.SeedSequence sklearn.feature_extraction.DictVectorizer sklearn.feature_selection.chi2 sklearn.feature_extraction.F

AI实战-营销活动和数据集分析预测实例（含19个源代码+523.50 KB完整的数据集）代码手工整理，无语法错误，可运行。包括：19个代码，共231.67 KB；数据大小：2个文件共523.50 KB。使用到的模块： pandas numpy seaborn matplotlib matplotlib.pyplot warnings sklearn.model_selection.train_test_split sklearn.ensemble.RandomForestClassifier sklearn.metrics.accuracy_score sklearn.metrics.confusion_matrix scipy.stats pingouin datetime.date matplotlib.ticker.PercentFormatter scipy.stats.f_oneway scipy.stats.friedmanchisquare sklearn.linear_model.LogisticRegression sklearn.metrics.precision_score sklearn.metrics.recall_score sklearn.metrics.f1_score sklearn.metrics.classification_report sklearn.metrics.roc_curve sklearn.metrics.roc_auc_score sklearn.svm sklearn.cluster.KMeans sklearn.preprocessing.StandardScaler sklearn.decomposition.PCA sklearn.metrics.silhouette_score sklearn.metrics.davies_bouldin_score sklearn.preprocessing.MinMaxScaler IPython.display.display sklearn.preprocessing.OneHotEncoder sklearn.model_selection.GridSearchCV sklearn.tree.DecisionTreeClassifier sklearn.tree.plot_tree sklearn.ensemble.GradientBoostingClassifier sklearn.metrics.RocCurveDisplay imblearn.over_sampling.SMOTE imblearn.pipeline.Pipeline sklearn.model_selection.RandomizedSearchCV sklearn.model_selection.StratifiedKFold sklearn.feature_selection.RFE sklearn.utils.class_weight sklearn.feature_selection.RFECV plotly.express datetime xgboost.XGBClassifier xgboost.plot_importance os sklearn.preprocessing.LabelEncoder kneed.KneeLocator sklearn.linear_model.LinearRegression sklearn.metrics.mean_squared_error sklearn.metrics.r2_score lightgbm.LGBMRegressor tensorflow sklearn.base.BaseEstimator sklearn.base.TransformerMixin sklearn.model_selection.cross_val_score sklearn.metrics.precision_recall_fscore_support xgboost datetime.datetime sklearn.metrics.ConfusionMatrixDisplay sklearn.tree.export_text missingno plotly.offline.init_notebook_mode plotly.io

AI实战-金融机构营销活动数据集分析预测实例（含20个源代码+897.42 KB完整的数据集）代码手工整理，无语法错误，可运行。包括：20个代码，共258.28 KB；数据大小：1个文件共897.42 KB。使用到的模块： matplotlib.pyplot numpy os pandas seaborn sklearn.preprocessing.LabelEncoder sklearn.model_selection.train_test_split sklearn.model_selection.RandomizedSearchCV sklearn.model_selection.GridSearchCV sklearn.ensemble.RandomForestClassifier sklearn.ensemble.GradientBoostingClassifier sklearn.tree.DecisionTreeClassifier xgboost.XGBClassifier sklearn.metrics.roc_curve sklearn.metrics.auc sklearn.metrics.accuracy_score warnings sklearn.svm.SVC sklearn.tree.plot_tree xgboost lightgbm sklearn.metrics.classification_report sklearn.metrics.roc_auc_score sklearn.metrics.confusion_matrix shap optuna sklearn.preprocessing.OneHotEncoder sklearn.preprocessing.RobustScaler sklearn.feature_selection.mutual_info_classif sklearn.cluster.KMeans sklearn.metrics.silhouette_score plotly.express plotly.graph_objects plotly.offline.init_notebook_mode plotly.io plotly.subplots.make_subplots sklearn.preprocessing.StandardScaler sklearn.linear_model.LogisticRegression sklearn.svm sklearn.neighbors.KNeighborsClassifier sklearn.naive_bayes.GaussianNB sklearn.model_selection.cross_val_score sklearn.tree sklearn.metrics.r2_score sklearn.inspection.permutation_importance sklearn.ensemble.AdaBoostClassifier sklearn.model_selection.validation_curve sklearn.metrics.precision_score sklearn.metrics.recall_score sklearn.metrics.f1_score sklearn.preprocessing.MinMaxScaler time.time warnings.simplefilter sklearn.decomposition.PCA kagglehub torch torch.nn torch.optim torch.nn.functional torch.utils.data.TensorDataset torch.utils.data.DataLoader time psutil plotly.tools plotly.figure_factory plotly.graph_objs plotly.offline.plot plotly.offline.iplot squarify sklearn.model_selection.StratifiedShuffleSplit sklearn.base.BaseEstimator sklearn.base.TransformerMixin sklearn.utils.check_array scipy.sparse sklearn.pipeline.Pipeline sklearn.pipeline.FeatureUnion sklearn.neural_network.MLPClassifier sklea

AI实战-世界世界各个城市的空气质量测量值数据集分析预测实例（含6个源代码+932.64 KB完整的数据集）代码手工整理，无语法错误，可运行。包括：6个代码，共34.40 KB；数据大小：2个文件共932.64 KB。使用到的模块： numpy pandas seaborn matplotlib.pyplot os sklearn.ensemble.IsolationForest sklearn.preprocessing.StandardScaler plotly.express plotly.graph_objects IPython.display.display sklearn.cluster.KMeans sklearn.metrics.silhouette_score sklearn.decomposition.PCA sklearn.model_selection.train_test_split sklearn.model_selection.cross_val_score sklearn.linear_model.LinearRegression sklearn.ensemble.RandomForestRegressor sklearn.metrics.mean_absolute_error sklearn.metrics.mean_squared_error sklearn.metrics.r2_score joblib warnings plotly.offline.init_notebook_mode IPython.display.IFrame plotly.offline.plot sklearn.preprocessing.LabelEncoder

AI实战-COVID 19世界疫苗接种数据分析预测实例（含18个源代码+18.34 MB完整的数据集）代码手工整理，无语法错误，可运行。包括：18个代码，共124.53 KB；数据大小：2个文件共18.34 MB。使用到的模块： os pandas matplotlib.pyplot seaborn plotly.express folium.features.Choropleth folium warnings numpy plotly.offline.iplot plotly.graph_objs plotly.io peakutils scipy.stats.mode plotly.offline.init_notebook_mode pandas_profiling.ProfileReport sklearn.preprocessing.StandardScaler sklearn.model_selection.train_test_split sklearn.model_selection.cross_val_score sklearn.model_selection.TimeSeriesSplit sklearn.model_selection.GridSearchCV sklearn.metrics.mean_squared_error sklearn.metrics.mean_absolute_error sklearn.metrics.r2_score xgboost lightgbm.LGBMRegressor torch tqdm.notebook.tqdm transformers.BertTokenizer torch.utils.data.TensorDataset transformers.BertForSequenceClassification torch.utils.data.DataLoader torch.utils.data.RandomSampler transformers.AdamW transformers.get_linear_schedule_with_warmup sklearn.metrics.f1_score random sklearn.linear_model.LinearRegression wordcloud.WordCloud wordninja spellchecker.SpellChecker collections.Counter nltk nltk.stem.WordNetLemmatizer nltk.corpus.stopwords sklearn.preprocessing.LabelEncoder sklearn.preprocessing.PolynomialFeatures sklearn.metrics sklearn.tree.DecisionTreeRegressor sklearn.cluster.KMeans missingno sklearn.metrics.confusion_matrix sklearn.metrics.roc_curve sklearn.metrics.accuracy_score sklearn.metrics.roc_auc_score imblearn.over_sampling.SMOTE sklearn.linear_model.LogisticRegression sklearn.ensemble.RandomForestClassifier sklearn.ensemble.GradientBoostingClassifier sklearn.tree.DecisionTreeClassifier catboost.CatBoostClassifier lightgbm.LGBMClassifier xgboost.XGBClassifier sklearn.ensemble.AdaBoostClassifier plotly.offline

37,743

社区成员

34,212

社区内容

发帖

与我相关

我的任务

社区管理员

加入社区

近7日
近30日
至今

加载中

查看更多榜单

试试用AI创作助手写篇文章吧

+ 用AI写文章