Machine learning 如何在聚类算法中使用sklearn.inspection.permutation\u重要性
在真正的问题中,我没有y(真正的标签),我试图做Machine learning 如何在聚类算法中使用sklearn.inspection.permutation\u重要性,machine-learning,scikit-learn,jupyter-notebook,cluster-analysis,feature-extraction,Machine Learning,Scikit Learn,Jupyter Notebook,Cluster Analysis,Feature Extraction,在真正的问题中,我没有y(真正的标签),我试图做y=None,使之成为一种无监督的学习。但它不起作用。我得到: import numpy as np from sklearn.datasets import make_classification from sklearn.cluster import KMeans X, y = make_classification(n_samples=1000, n_features=4,
y=None
,使之成为一种无监督的学习。但它不起作用。我得到:
import numpy as np
from sklearn.datasets import make_classification
from sklearn.cluster import KMeans
X, y = make_classification(n_samples=1000,
n_features=4,
n_informative=3,
n_redundant=0,
n_repeated=0,
n_classes=2,
random_state=0,
shuffle=False)
km = KMeans(n_clusters=3).fit(X)
result = permutation_importance(km, X, y, scoring='homogeneity_score', n_repeats=10, random_state=0, n_jobs=-1)
result
---------------------------------------------------------------------------
ValueError回溯(最近一次调用上次)
在()
---->1结果=排列重要性(km,X,y=无,评分=”同质性评分',n次重复=10,随机状态=0,n次作业=-1)
5帧
/usr/local/lib/python3.6/dist-packages/sklearn/metrics/cluster//u supervised.py in check\u clusterings(labels\u true,labels\u pred)
53如果标签_true.ndim!=1:
54升值误差(
--->55“labels\u true必须为1D:形状为%r”%(labels\u true.shape,)
56如果标签_pred.ndim!=1:
57升值误差(
ValueError:labels\u true必须为1D:形状为()
有人知道如何在没有真标签的情况下实现吗?首先,证明k-means对特征的排列是不变的很简单……因为和是排列不变的 如果您仍然想进行实验,请尝试使用0数组作为y
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-72-81045ae9cb66> in <module>()
----> 1 result = permutation_importance(km, X, y=None, scoring='homogeneity_score', n_repeats=10, random_state=0, n_jobs=-1)
5 frames
/usr/local/lib/python3.6/dist-packages/sklearn/metrics/cluster/_supervised.py in check_clusterings(labels_true, labels_pred)
53 if labels_true.ndim != 1:
54 raise ValueError(
---> 55 "labels_true must be 1D: shape is %r" % (labels_true.shape,))
56 if labels_pred.ndim != 1:
57 raise ValueError(
ValueError: labels_true must be 1D: shape is ()