Spaces:
Sleeping
Sleeping
File size: 2,124 Bytes
3cab2dd 932b3cb 3cab2dd 932b3cb 3cab2dd c2687ce 3cab2dd fcc4124 c2687ce fcc4124 803aad3 c2687ce 803aad3 fcc4124 3cab2dd 527f863 c2687ce 527f863 28d5b3d c2687ce |
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 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 |
from dataclasses import dataclass
import numpy as np
from cluster.clusterer import Clusterer
@dataclass
class Kmeans(Clusterer):
k: int
max_iter: int
centroids = None
clusters = None
def build(
self,
X: np.array,
) -> None:
# randomly initialize centroids
centroids = X[np.random.choice(
X.shape[0],
self.k,
replace=False,
)]
# Calculate Euclidean distance between each data point and each centroid
# then assign each point to its closest cluster
clusters = self.assign_clusters(X, centroids)
centroids = self.update_centroids(self.k, X, clusters)
while True:
new_clusts = self.assign_clusters(X, centroids)
if np.array_equal(new_clusts, clusters):
break
clusters = new_clusts
centroids = self.update_centroids(self.k, X, clusters)
self.clusters = clusters
self.centroids = centroids
@staticmethod
def assign_clusters(
X: np.array,
centroids: np.array,
) -> np.array:
distances = np.sqrt(((X - centroids[:, np.newaxis])**2).sum(axis=2))
clusts = np.argmin(distances, axis=0)
return clusts
@staticmethod
def update_centroids(
k: int,
X: np.array,
clusters: np.array,
) -> np.array:
centroids = np.zeros((k, X.shape[1]))
for i in range(k):
centroids[i] = X[clusters == i].mean(axis=0)
return centroids
def to_dict(
self,
X: np.array,
) -> dict:
cluster_data = []
for i in range(self.k):
indices = np.where(self.clusters == i)[0]
cluster_pts = X[indices].tolist()
cluster_data.append({
"cluster_id": i,
"centroid": self.centroids[i].tolist(),
"points": cluster_pts,
})
return {
"k": self.k,
"max_iter": self.max_iter,
"clusters": cluster_data,
"plot_key": self.plot_key,
}
|