当我绘制特征重要性时，得到的是一个混乱的图表。我有超过7000个变量。我知道内置函数只选择最重要的变量，但最终的图表难以阅读。这是我完整的代码:
import numpy as npimport pandas as pddf = pd.read_csv('ricerice.csv')array=df.valuesX = array[:,0:7803]Y = array[:,7804]from xgboost import XGBClassifierfrom sklearn.model_selection import train_test_splitfrom sklearn.metrics import accuracy_scoreseed=0test_size=0.30X_train, X_test, y_train, y_test = train_test_split(X,Y,test_size=test_size, random_state=seed)from xgboost import XGBClassifiermodel = XGBClassifier()model.fit(X, Y)import matplotlib.pyplot as pltfrom matplotlib import pyplotfrom xgboost import plot_importancefig1=plt.gcf()plot_importance(model)plt.draw()fig1.savefig('xgboost.png', figsize=(50, 40), dpi=1000)尽管图表尺寸很大，但图表仍然难以辨认。 
回答：
有几个要点需要注意:
- 要拟合模型，你应该使用训练数据集(
X_train, y_train)，而不是整个数据集(X, y)。 - 你可以使用
plot_importance()函数的max_num_features参数来只显示前max_num_features个特征（例如前10个）。
通过对你的代码进行上述修改，并使用一些随机生成的数据，代码和输出如下所示:
import numpy as np# generate some random data for demonstration purpose, use your original dataset hereX = np.random.rand(1000,100) # 1000 x 100 datay = np.random.rand(1000).round() # 0, 1 labelsfrom xgboost import XGBClassifierfrom sklearn.model_selection import train_test_splitfrom sklearn.metrics import accuracy_scoreseed=0test_size=0.30X_train, X_test, y_train, y_test = train_test_split(X,y,test_size=test_size, random_state=seed)from xgboost import XGBClassifiermodel = XGBClassifier()model.fit(X_train, y_train)import matplotlib.pylab as pltfrom matplotlib import pyplotfrom xgboost import plot_importanceplot_importance(model, max_num_features=10) # top 10 most important featuresplt.show()
