daily pastebin goal
63%
SHARE
TWEET

Untitled

a guest May 17th, 2018 91 Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
  1. for i in range(4):
  2.         y_train= y[:,i]
  3.         print('Train subject %d, class %s' % (subject, cols[i]))
  4.         lr.fit(X_train[::sample,:],y_train[::sample])
  5.         pred[:,i] = lr.predict_proba(X_test)[:,1]
  6.    
  7. prediction = lr.predict(X_train)
  8.  
  9. print(confusion_matrix(y_train, prediction))
  10.    
  11. import numpy as np
  12.  
  13.  
  14. def plot_confusion_matrix(cm,
  15.                           target_names,
  16.                           title='Confusion matrix',
  17.                           cmap=None,
  18.                           normalize=True):
  19.     """
  20.     given a sklearn confusion matrix (cm), make a nice plot
  21.  
  22.     Arguments
  23.     ---------
  24.     cm:           confusion matrix from sklearn.metrics.confusion_matrix
  25.  
  26.     target_names: given classification classes such as [0, 1, 2]
  27.                   the class names, for example: ['high', 'medium', 'low']
  28.  
  29.     title:        the text to display at the top of the matrix
  30.  
  31.     cmap:         the gradient of the values displayed from matplotlib.pyplot.cm
  32.                   see http://matplotlib.org/examples/color/colormaps_reference.html
  33.                   plt.get_cmap('jet') or plt.cm.Blues
  34.  
  35.     normalize:    If False, plot the raw numbers
  36.                   If True, plot the proportions
  37.  
  38.     Usage
  39.     -----
  40.     plot_confusion_matrix(cm           = cm,                  # confusion matrix created by
  41.                                                               # sklearn.metrics.confusion_matrix
  42.                           normalize    = True,                # show proportions
  43.                           target_names = y_labels_vals,       # list of names of the classes
  44.                           title        = best_estimator_name) # title of graph
  45.  
  46.     Citiation
  47.     ---------
  48.     http://scikit-learn.org/stable/auto_examples/model_selection/plot_confusion_matrix.html
  49.  
  50.     """
  51.     import matplotlib.pyplot as plt
  52.     import numpy as np
  53.     import itertools
  54.  
  55.     accuracy = np.trace(cm) / float(np.sum(cm))
  56.     misclass = 1 - accuracy
  57.  
  58.     if cmap is None:
  59.         cmap = plt.get_cmap('Blues')
  60.  
  61.     plt.figure(figsize=(8, 6))
  62.     plt.imshow(cm, interpolation='nearest', cmap=cmap)
  63.     plt.title(title)
  64.     plt.colorbar()
  65.  
  66.     if target_names is not None:
  67.         tick_marks = np.arange(len(target_names))
  68.         plt.xticks(tick_marks, target_names, rotation=45)
  69.         plt.yticks(tick_marks, target_names)
  70.  
  71.     if normalize:
  72.         cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]
  73.  
  74.  
  75.     thresh = cm.max() / 1.5 if normalize else cm.max() / 2
  76.     for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):
  77.         if normalize:
  78.             plt.text(j, i, "{:0.4f}".format(cm[i, j]),
  79.                      horizontalalignment="center",
  80.                      color="white" if cm[i, j] > thresh else "black")
  81.         else:
  82.             plt.text(j, i, "{:,}".format(cm[i, j]),
  83.                      horizontalalignment="center",
  84.                      color="white" if cm[i, j] > thresh else "black")
  85.  
  86.  
  87.     plt.tight_layout()
  88.     plt.ylabel('True label')
  89.     plt.xlabel('Predicted labelnaccuracy={:0.4f}; misclass={:0.4f}'.format(accuracy, misclass))
  90.     plt.show()
RAW Paste Data
Top