■ k-평균 클러스터링(k-means clustering)을 사용하는 방법을 보여준다.
▶ 예제 코드 (PY)
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 |
import sklearn.cluster as cluster import sklearn.datasets as datasets import sklearn.metrics as metrics mnistBunch = datasets.load_digits() imageNDArray = mnistBunch.images imageCount = len(imageNDArray) imageNDArray = imageNDArray.reshape(len(imageNDArray), -1) labelNDArray = mnistBunch.target trainCount = int((imageCount / 4) * 3) testCount = int((imageCount / 4)) kmeans = cluster.KMeans(n_clusters = 10, init = "k-means++", n_init = 10) kmeans.fit(imageNDArray[:trainCount]) print(kmeans.labels_) print(type(kmeans.labels_)) print(kmeans.labels_.shape) print() testLabelList = labelNDArray[testCount:] predictionLabelList = kmeans.predict(imageNDArray[testCount:]) print("성능 리포트 : \n %s \n" % (metrics.classification_report(testLabelList, predictionLabelList))) """ [6 4 4 ... 3 1 3] <class 'numpy.ndarray'> (1347,) 성능 리포트 : precision recall f1-score support 0 0.00 0.00 0.00 131 1 0.01 0.01 0.01 137 2 0.00 0.00 0.00 131 3 0.85 0.82 0.83 136 4 0.00 0.00 0.00 139 5 0.91 0.72 0.80 136 6 0.01 0.01 0.01 138 7 0.85 0.98 0.91 134 8 0.02 0.02 0.02 130 9 0.00 0.00 0.00 136 accuracy 0.26 1348 macro avg 0.27 0.26 0.26 1348 weighted avg 0.27 0.26 0.26 1348 """ |