python画ROC曲线如何画的好看

import numpy as np
from sklearn.metrics import confusion_matrix, accuracy_score,f1_score,roc_auc_score,recall_score,precision_score
from sklearn import metrics
from matplotlib import pyplot


styles=['fivethirtyeight',
 'dark_background',
 'bmh',
 'classic',
 'seaborn-dark',
 'grayscale',
 'seaborn-deep',
 'seaborn-muted',
 'seaborn-colorblind',
 'seaborn-white',
 'seaborn-dark-palette',
 'ggplot',
 'tableau-colorblind10',
 '_classic_test',
 'seaborn-darkgrid',
 'seaborn-notebook',
 'Solarize_Light2',
 'seaborn-paper',
 'seaborn-whitegrid',
 'seaborn-pastel',
 'seaborn-talk',
 'seaborn-bright',
 'seaborn',
 'seaborn-ticks',
 'seaborn-poster',
 'fast']

def rocs_plot(y2, P2, setname=''):

    pyplot.figure(figsize=(5, 4), dpi=100)
    pyplot.style.use('seaborn-darkgrid')
    
    palette = pyplot.get_cmap('Set1')
    
        
    for i, p2 in enumerate(P2):
        fpr1, tpr1, thresholds1 = metrics.roc_curve(y2[i], p2, pos_label=1)
        roc_auc1 = metrics.auc(fpr1, tpr1)   
        pyplot.plot(fpr1, tpr1, lw=1, color=palette(i), linewidth=1.5, alpha=0.9, label=result_paths[i][:-10]+", AUC=%0.3f" % (roc_auc1))       
    pyplot.xlim([0.00, 1.0])
    pyplot.ylim([0.00, 1.0])
    pyplot.xlabel("1-Specificity",fontsize=12)
    pyplot.ylabel("Sensitivity",fontsize=12)
    pyplot.title("ROC",fontsize=12)
    pyplot.legend(loc="lower right")
    pyplot.savefig('ROC.png',dpi=300)

pyplot画图太丑，改个风格就好

pyplot.style.use('seaborn-darkgrid')

里面的风格可以自己改，'seaborn-darkgrid' 风格出来大概是这样的，比默认格式好看多了：