more PEP8

scikits/learn/metrics.py

@@ -1,9 +1,12 @@
+"""Utilities to evaluate the predictive performance of models"""
+
 # Authors: Alexandre Gramfort <[email protected]>
 # Mathieu Blondel <[email protected]>
 # License: BSD Style.
 
 import numpy as np
 
+
 def confusion_matrix(y, y_):
  """Compute confusion matrix to evaluate the accuracy of a classification
 
@@ -38,10 +41,10 @@ def confusion_matrix(y, y_):
  labels = np.unique(labels)
  n_labels = labels.size
 
- cm = np.empty((n_labels,n_labels))
+ cm = np.empty((n_labels, n_labels))
  for i, label_i in enumerate(labels):
  for j, label_j in enumerate(labels):
- cm[i,j] = np.sum(np.logical_and(y==label_i, y_==label_j))
+ cm[i, j] = np.sum(np.logical_and(y == label_i, y_ == label_j))
 
  return cm
 
@@ -77,13 +80,15 @@ def roc_curve(y, probas_):
  probas_ = probas_.ravel()
  thresholds = np.sort(np.unique(probas_))[::-1]
  n_thresholds = thresholds.size
+
  tpr = np.empty(n_thresholds) # True positive rate
  fpr = np.empty(n_thresholds) # False positive rate
- n_pos = float(np.sum(y==1)) # nb of true positive
- n_neg = float(np.sum(y==0)) # nb of true negative
+ n_pos = float(np.sum(y == 1)) # nb of true positive
+ n_neg = float(np.sum(y == 0)) # nb of true negative
+
  for i, t in enumerate(thresholds):
- tpr[i] = np.sum(y[probas_>=t]==1) / n_pos
- fpr[i] = np.sum(y[probas_>=t]==0) / n_neg
+ tpr[i] = np.sum(y[probas_ >= t] == 1) / n_pos
+ fpr[i] = np.sum(y[probas_ >= t] == 0) / n_neg
 
  return fpr, tpr, thresholds
 
@@ -134,8 +139,8 @@ def precision(y_true, y_pred):
  =======
  precision : float
  """
- true_pos = np.sum(y_true[y_pred == 1]==1)
- false_pos = np.sum(y_true[y_pred == 1]==0)
+ true_pos = np.sum(y_true[y_pred == 1] == 1)
+ false_pos = np.sum(y_true[y_pred == 1] == 0)
  return true_pos / float(true_pos + false_pos)
 
 
@@ -160,8 +165,8 @@ def recall(y_true, y_pred):
  =======
  recall : float
  """
- true_pos = np.sum(y_true[y_pred == 1]==1)
- false_neg = np.sum(y_true[y_pred == 0]==1)
+ true_pos = np.sum(y_true[y_pred == 1] == 1)
+ false_neg = np.sum(y_true[y_pred == 0] == 1)
  return true_pos / float(true_pos + false_neg)
 
 
@@ -194,9 +199,9 @@ def precision_recall(y_true, y_pred):
  ==========
  http:https://en.wikipedia.org/wiki/Precision_and_recall
  """
- true_pos = np.sum(y_true[y_pred == 1]==1)
- false_pos = np.sum(y_true[y_pred == 1]==0)
- false_neg = np.sum(y_true[y_pred == 0]==1)
+ true_pos = np.sum(y_true[y_pred == 1] == 1)
+ false_pos = np.sum(y_true[y_pred == 1] == 0)
+ false_neg = np.sum(y_true[y_pred == 0] == 1)
  precision = true_pos / float(true_pos + false_pos)
  recall = true_pos / float(true_pos + false_neg)
  return precision, recall