make_blobs函数

示例1：

# make_blobs示例
from sklearn.datasets.samples_generator import make_blobs

X, y = make_blobs(n_samples=10, centers=3, n_features=2,
                   random_state=0)
#看看数据集长什么样
plt.scatter(X[:, 0], X[:, 1], c=y, cmap="rainbow");

示例2：

X, y = make_blobs(n_samples=[3, 3, 4], centers=None, n_features=2,
                  random_state=0)

#看看数据集长什么样
plt.scatter(X[:, 0], X[:, 1], c=y, cmap="rainbow");

可见，n_samples=10与n_samples=[3, 3, 4]等价，centers=3与centers=None等价，验证了默认值的定义。

示例3：

# 创建数据集
class_1_ = 7
class_2_ = 4
centers_ = [[0.0, 0.0], [1,1]]
clusters_std = [0.5, 1]
X_, y_ = make_blobs(n_samples=[class_1_, class_2_],
                    centers=centers_,
                    cluster_std=clusters_std,
                    random_state=0, shuffle=False)

# or :
X_, y_ = make_blobs(n_samples=[7, 4],
                    centers=[[0.0, 0.0], [1,1]],
                    cluster_std=[0.5, 1],
                    random_state=0, shuffle=False)    # n_features 默认是2

# 绘图：
plt.scatter(X_[:, 0], X_[:, 1], c=y_, cmap="rainbow",s=100); # s控制点的大小