CODEWITHSUNDEEP
scikit-learnk-means
tomjirsa
tomjirsa

Reputation: 11

sklearn - KMeans of arrays: ValueError: setting an array element with a sequence

I am trying to do KMeans Clusterin over multidimensional features. I get ValueError: setting an array element with a sequence.

Here is an example of what I have already tried:

import pandas as pd
from sklearn.cluster import KMeans

test = pd.DataFrame(np.random.randint(low=0, high=10, size=(30, 4)), columns=['a', 'b', 'c', 'd'])
test["combined1"] = test.loc(axis=1)["a","b"].values.tolist()
test["combined2"] = test.loc(axis=1)["c","d"].values.tolist()
test.drop(['a', 'b', 'c', 'd'],axis=1, inplace=True)
test.head()

kmeans = KMeans(n_clusters=3, random_state=0)
kmeans.fit(test)

The KMeans fit fails with

/usr/local/lib/python3.5/dist-packages/numpy/core/numeric.py in asarray(a, dtype, order)
    490 
    491     """
--> 492     return array(a, dtype, copy=False, order=order)
    493 
    494 

ValueError: setting an array element with a sequence.

Upvotes: 1

Views: 1650

Answers (1)

andrewchauzov
andrewchauzov

Reputation: 1009

So, you pass sequences into KMeans (like [8, 1]) and that's why it does not work. Please check here:

https://scikit-learn.org/stable/modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans.fit

that fit() method allows you to use:

X : array-like or sparse matrix, shape=(n_samples, n_features)

Upvotes: 1

Related Questions