我使用了來自 Sklearn 的 Digits 數(shù)據(jù)集，并嘗試將維度從 64 減少到 2：import numpy as npimport pandas as pdimport matplotlib.pyplot as pltimport seaborn as sns#%matplotib inlinefrom sklearn.decomposition import PCAfrom sklearn.datasets import load_digitsdigits = load_digits()digits_df = pd.DataFrame(digits.data,)digits_df["target"] = pd.Series(digits.target)pca = PCA(n_components=2)digits_pca = pca.fit_transform(digits_df.iloc[:,:64])digits_df_pca = pd.DataFrame(digits_pca, columns =["Component1","Component2"])finalDf = pd.concat([digits_df_pca, digits_df["target"]], axis = 1)plt.figure(figsize=(10,10))sns.scatterplot(data=finalDf,x="Component1", y = "Component2",hue="target", )圖表：圖中唯一的數(shù)字是 0、3、6、9。為什么我看不到其他五個數(shù)字？
1 回答
互換的青春
TA貢獻1797條經(jīng)驗 獲得超6個贊
如果所有標簽都可用一組標簽，則簽入您的數(shù)據(jù)。
如果是，那么您可以嘗試
sns.scatterplot(data=finalDf,x="Component1", y = "Component2",hue="target",
legend = 'full')
工作代碼：
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
#%matplotib inline
from sklearn.decomposition import PCA
from sklearn.datasets import load_digits
digits = load_digits()
digits_df = pd.DataFrame(digits.data,)
digits_df["target"] = pd.Series(digits.target)
pca = PCA(n_components=2)
digits_pca = pca.fit_transform(digits_df.iloc[:,:64])
digits_df_pca = pd.DataFrame(digits_pca,
columns =["Component1","Component2"])
finalDf = pd.concat([digits_df_pca, digits_df["target"]], axis = 1)
plt.figure(figsize=(10,10))
palette = sns.color_palette("bright", 10)
sns.scatterplot(data=finalDf,x="Component1", y = "Component2",hue="target",
legend = 'full', palette = palette)
添加回答
舉報
0/150
提交
取消