umap plot

umap库其实可以自己画图，就不用每次自己调用matplotlib或者scanpy进行画图了

试验如下

import sklearn.datasets
import pandas as pd
import numpy as np
import umap

pendigits = sklearn.datasets.load_digits()
mnist = sklearn.datasets.fetch_openml('mnist_784')
fmnist = sklearn.datasets.fetch_openml('Fashion-MNIST')

mapper = umap.UMAP().fit(pendigits.data)

import umap.plot

umap.plot.points(mapper)

umap.plot.points(mapper, labels=pendigits.target)

umap.plot.points(mapper, values=pendigits.data.mean(axis=1), theme='fire')

umap.plot.points(mapper, labels=pendigits.target, color_key_cmap='Paired', background='black')

结果如下

保存图片

     mapper = umap.UMAP().fit(emb)
     p = umap.plot.points(mapper, labels=target)
     plt.savefig("./aa.png")