diff --git a/scikits/learn/metrics.py b/scikits/learn/metrics.py
index c04582dac9ca97fea58be3c6c1e2896b26f78791..1b3ac7762783db61c36fd66e72ec6d78de221e66 100644
--- a/scikits/learn/metrics.py
+++ b/scikits/learn/metrics.py
@@ -155,7 +155,7 @@ def precision(y_true, y_pred):
=======
precision : float
"""
- return precision_recall_fscore(y_true, y_pred)[0]
+ return precision_recall_fscore_support(y_true, y_pred)[0]
def recall(y_true, y_pred):
@@ -179,7 +179,7 @@ def recall(y_true, y_pred):
=======
recall : array, shape = [n_unique_labels], dtype = np.double
"""
- return precision_recall_fscore(y_true, y_pred)[1]
+ return precision_recall_fscore_support(y_true, y_pred)[1]
def fbeta_score(y_true, y_pred, beta):
@@ -241,8 +241,8 @@ def f1_score(y_true, y_pred):
return fbeta_score(y_true, y_pred, 1)
-def precision_recall_fscore(y_true, y_pred, beta=1.0, labels=None):
- """Compute precision and recall and f-measure at the same time.
+def precision_recall_fscore_support(y_true, y_pred, beta=1.0, labels=None):
+ """Compute precisions, recalls, f-measures and support for each class
The precision is the ratio :math:`tp / (tp + fp)` where tp is the number of
true positives and fp the number of false positives. The precision is
@@ -253,11 +253,14 @@ def precision_recall_fscore(y_true, y_pred, beta=1.0, labels=None):
true positives and fn the number of false negatives. The recall is
intuitively the ability of the classifier to find all the positive samples.
- The F_beta score can be interpreted as a weighted average of the precision
- and recall, where an F_beta score reaches its best value at 1 and worst
- score at 0.
+ The F_beta score can be interpreted as a weighted harmonic mean of
+ the precision and recall, where an F_beta score reaches its best
+ value at 1 and worst score at 0.
- The F_1 score weights recall beta as much as precision.
+ The F_beta score weights recall beta as much as precision. beta = 1.0 means
+ recall and precsion are as important.
+
+ The support is the number of occurrences of each class in y_true.
Parameters
==========
@@ -274,7 +277,8 @@ def precision_recall_fscore(y_true, y_pred, beta=1.0, labels=None):
=======
precision: array, shape = [n_unique_labels], dtype = np.double
recall: array, shape = [n_unique_labels], dtype = np.double
- precision: array, shape = [n_unique_labels], dtype = np.double
+ f1_score: array, shape = [n_unique_labels], dtype = np.double
+ support: array, shape = [n_unique_labels], dtype = np.long
References
==========
@@ -290,11 +294,13 @@ def precision_recall_fscore(y_true, y_pred, beta=1.0, labels=None):
true_pos = np.zeros(n_labels, dtype=np.double)
false_pos = np.zeros(n_labels, dtype=np.double)
false_neg = np.zeros(n_labels, dtype=np.double)
+ support = np.zeros(n_labels, dtype=np.long)
for i, label_i in enumerate(labels):
true_pos[i] = np.sum(y_pred[y_true == label_i] == label_i)
false_pos[i] = np.sum(y_pred[y_true != label_i] == label_i)
false_neg[i] = np.sum(y_pred[y_true == label_i] != label_i)
+ support[i] = np.sum(y_true == label_i)
# precision and recall
precision = true_pos / (true_pos + false_pos)
@@ -304,7 +310,7 @@ def precision_recall_fscore(y_true, y_pred, beta=1.0, labels=None):
beta2 = beta ** 2
fscore = (1 + beta2) * (precision * recall) / (
beta2 * precision + recall)
- return precision, recall, fscore
+ return precision, recall, fscore, support
def classification_report(y_true, y_pred, labels=None, class_names=None):
@@ -336,15 +342,17 @@ def classification_report(y_true, y_pred, labels=None, class_names=None):
else:
labels = np.asarray(labels, dtype=np.int)
+ last_line_heading = 'avg / total'
+
if class_names is None:
- width = len('mean')
+ width = len(last_line_heading)
class_names = ['%d' % l for l in labels]
else:
width = max(len(cn) for cn in class_names)
- width = max(width, len('mean'))
+ width = max(width, len(last_line_heading))
- headers = ["precision", "recall", "f1-score"]
+ headers = ["precision", "recall", "f1-score", "support"]
fmt = '{0:>%d}' % width # first column: class name
fmt += ' '
fmt += ' '.join(['{%d:>9}' % (i + 1) for i, _ in enumerate(headers)])
@@ -354,19 +362,23 @@ def classification_report(y_true, y_pred, labels=None, class_names=None):
report = fmt.format(*headers)
report += '\n'
- p, r, f1 = precision_recall_fscore(y_true, y_pred, labels=labels)
+ p, r, f1, s = precision_recall_fscore_support(y_true, y_pred, labels=labels)
for i, label in enumerate(labels):
values = [class_names[i]]
for v in (p[i], r[i], f1[i]):
values += ["%0.2f" % float(v)]
+ values += ["%d" % int(s[i])]
report += fmt.format(*values)
report += '\n'
# compute averages
- values = ['mean']
- for v in (np.mean(p), np.mean(r), np.mean(f1)):
+ values = [last_line_heading]
+ for v in (np.average(p, weights=s),
+ np.average(r, weights=s),
+ np.average(f1, weights=s)):
values += ["%0.2f" % float(v)]
+ values += ['%d' % np.sum(s)]
report += fmt.format(*values)
return report
@@ -421,7 +433,7 @@ def precision_recall_curve(y_true, probas_pred):
for i, t in enumerate(thresholds):
y_pred = np.ones(len(y_true))
y_pred[probas_pred < t] = 0
- p, r, _ = precision_recall_fscore(y_true, y_pred)
+ p, r, _, _ = precision_recall_fscore_support(y_true, y_pred)
precision[i] = p[1]
recall[i] = r[1]
precision[-1] = 1.0
diff --git a/scikits/learn/tests/test_metrics.py b/scikits/learn/tests/test_metrics.py
index d5509f439282551a008cbb2f7b648c0f996528c2..4810405a30697122df378474c87b27d510b1bbbc 100644
--- a/scikits/learn/tests/test_metrics.py
+++ b/scikits/learn/tests/test_metrics.py
@@ -17,7 +17,7 @@ from ..metrics import f1_score
from ..metrics import mean_square_error
from ..metrics import precision
from ..metrics import precision_recall_curve
-from ..metrics import precision_recall_fscore
+from ..metrics import precision_recall_fscore_support
from ..metrics import recall
from ..metrics import roc_curve
from ..metrics import zero_one
@@ -80,10 +80,11 @@ def test_precision_recall_f1_score_binary():
"""Test Precision Recall and F1 Score for binary classification task"""
y_true, y_pred, _ = make_prediction(binary=True)
- p, r, f = precision_recall_fscore(y_true, y_pred)
+ p, r, f, s = precision_recall_fscore_support(y_true, y_pred)
assert_array_almost_equal(p, [0.73, 0.75], 2)
assert_array_almost_equal(r, [0.76, 0.72], 2)
assert_array_almost_equal(f, [0.75, 0.74], 2)
+ assert_array_equal(s, [25, 25])
def test_confusion_matrix_binary():
@@ -99,16 +100,19 @@ def test_precision_recall_f1_score_multiclass():
y_true, y_pred, _ = make_prediction(binary=False)
# compute scores with default labels introspection
- p, r, f = precision_recall_fscore(y_true, y_pred)
+ p, r, f, s = precision_recall_fscore_support(y_true, y_pred)
assert_array_almost_equal(p, [0.82, 0.55, 0.47], 2)
assert_array_almost_equal(r, [0.92, 0.17, 0.90], 2)
assert_array_almost_equal(f, [0.87, 0.26, 0.62], 2)
+ assert_array_equal(s, [25, 30, 20])
# same prediction but with and explicit label ordering
- p, r, f = precision_recall_fscore(y_true, y_pred, labels=[0, 2, 1])
+ p, r, f, s = precision_recall_fscore_support(
+ y_true, y_pred, labels=[0, 2, 1])
assert_array_almost_equal(p, [0.82, 0.47, 0.55], 2)
assert_array_almost_equal(r, [0.92, 0.90, 0.17], 2)
assert_array_almost_equal(f, [0.87, 0.62, 0.26], 2)
+ assert_array_equal(s, [25, 20, 30])
def test_confusion_matrix_multiclass():
@@ -135,13 +139,13 @@ def test_classification_report():
# print classification report with class names
expected_report = """\
- precision recall f1-score
+ precision recall f1-score support
- setosa 0.82 0.92 0.87
-versicolor 0.56 0.17 0.26
- virginica 0.47 0.90 0.62
+ setosa 0.82 0.92 0.87 25
+ versicolor 0.56 0.17 0.26 30
+ virginica 0.47 0.90 0.62 20
- mean 0.62 0.66 0.58
+avg / total 0.62 0.61 0.56 75
"""
report = classification_report(
y_true, y_pred, labels=range(len(iris.target_names)),
@@ -150,13 +154,13 @@ versicolor 0.56 0.17 0.26
# print classification report with label detection
expected_report = """\
- precision recall f1-score
+ precision recall f1-score support
- 0 0.82 0.92 0.87
- 1 0.56 0.17 0.26
- 2 0.47 0.90 0.62
+ 0 0.82 0.92 0.87 25
+ 1 0.56 0.17 0.26 30
+ 2 0.47 0.90 0.62 20
-mean 0.62 0.66 0.58
+avg / total 0.62 0.61 0.56 75
"""
report = classification_report(y_true, y_pred)
assert_equal(report, expected_report)