如何从 GridSearchCV 绘制网格分数? [英] How to graph grid scores from GridSearchCV?

问题描述

我正在寻找一种在 sklearn 中从 GridSearchCV 绘制 grid_scores_ 的方法.在这个例子中，我试图网格搜索 SVR 算法的最佳 gamma 和 C 参数.我的代码如下所示:

 C_range = 10.0 ** np.arange(-4, 4)gamma_range = 10.0 ** np.arange(-4, 4)param_grid = dict(gamma=gamma_range.tolist(), C=C_range.tolist())网格 = GridSearchCV(SVR(kernel='rbf', gamma=0.1),param_grid, cv=5)grid.fit(X_train,y_train)打印(grid.grid_scores_)

运行代码并打印网格分数后，我得到以下结果:

[mean: -3.28593, std: 1.69134, params: {'gamma': 0.0001, 'C': 0.0001}, mean: -3.29370, std: 1.69346, params: {'gamma', 0.001'C':0.0001}，均值:-3.28933，标准:1.69104，参数:{'gamma':0.01，'C':0.0001}，均值:-3.28925，标准:1.69106，参数:{'，gamma':0'C':0.0001}，均值:-3.28925，标准:1.69106，参数:{'gamma':1.0，'C':0.0001}，均值:-3.28925，标准:1.69106，参数:{'gamma':10'C':0.0001} 等]

我想根据 gamma 和 C 参数可视化所有分数(平均值).我试图获得的图表应如下所示:

其中 x 轴是 gamma，y 轴是平均得分(本例中的均方根误差)，不同的线代表不同的 C 值.

解决方案

from sklearn.svm import SVC从 sklearn.grid_search 导入 GridSearchCV从 sklearn 导入数据集导入 matplotlib.pyplot 作为 plt将 seaborn 作为 sns 导入将 numpy 导入为 np数字 = datasets.load_digits()X = 数字.数据y = 数字.目标clf_ = SVC(kernel='rbf')Cs = [1, 10, 100, 1000]伽马 = [1e-3, 1e-4]clf = GridSearchCV(clf_,dict(C=Cs,伽马=伽马)，cv=2，pre_dispatch='1*n_jobs',n_jobs=1)clf.fit(X, y)score = [x[1] for x in clf.grid_scores_]分数 = np.array(scores).reshape(len(Cs), len(Gammas))对于 ind，我在 enumerate(Cs) 中:plt.plot(Gammas, score[ind], label='C:' + str(i))plt.legend()plt.xlabel('Gamma')plt.ylabel('平均分数')plt.show()

• 代码基于

  I am looking for a way to graph grid_scores_ from GridSearchCV in sklearn. In this example I am trying to grid search for best gamma and C parameters for an SVR algorithm. My code looks as follows:

      C_range = 10.0 ** np.arange(-4, 4)
      gamma_range = 10.0 ** np.arange(-4, 4)
      param_grid = dict(gamma=gamma_range.tolist(), C=C_range.tolist())
      grid = GridSearchCV(SVR(kernel='rbf', gamma=0.1),param_grid, cv=5)
      grid.fit(X_train,y_train)
      print(grid.grid_scores_)
  

  After I run the code and print the grid scores I get the following outcome:

  [mean: -3.28593, std: 1.69134, params: {'gamma': 0.0001, 'C': 0.0001}, mean: -3.29370, std: 1.69346, params: {'gamma': 0.001, 'C': 0.0001}, mean: -3.28933, std: 1.69104, params: {'gamma': 0.01, 'C': 0.0001}, mean: -3.28925, std: 1.69106, params: {'gamma': 0.1, 'C': 0.0001}, mean: -3.28925, std: 1.69106, params: {'gamma': 1.0, 'C': 0.0001}, mean: -3.28925, std: 1.69106, params: {'gamma': 10.0, 'C': 0.0001},etc] 
  

  I would like to visualize all the scores (mean values) depending on gamma and C parameters. The graph I am trying to obtain should look as follows:

  Where x-axis is gamma, y-axis is mean score (root mean square error in this case), and different lines represent different C values.

  解决方案

  from sklearn.svm import SVC
  from sklearn.grid_search import GridSearchCV
  from sklearn import datasets
  import matplotlib.pyplot as plt
  import seaborn as sns
  import numpy as np
  
  digits = datasets.load_digits()
  X = digits.data
  y = digits.target
  
  clf_ = SVC(kernel='rbf')
  Cs = [1, 10, 100, 1000]
  Gammas = [1e-3, 1e-4]
  clf = GridSearchCV(clf_,
              dict(C=Cs,
                   gamma=Gammas),
                   cv=2,
                   pre_dispatch='1*n_jobs',
                   n_jobs=1)
  
  clf.fit(X, y)
  
  scores = [x[1] for x in clf.grid_scores_]
  scores = np.array(scores).reshape(len(Cs), len(Gammas))
  
  for ind, i in enumerate(Cs):
      plt.plot(Gammas, scores[ind], label='C: ' + str(i))
  plt.legend()
  plt.xlabel('Gamma')
  plt.ylabel('Mean score')
  plt.show()
  

  • Code is based on this.
  • Only puzzling part: will sklearn always respect the order of C & Gamma -> official example uses this "ordering"

  Output:

  这篇关于如何从 GridSearchCV 绘制网格分数?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！