log_loss#
- sklearn.metrics.log_loss(y_true,y_pred,*,normalize=True,sample_weight=None,labels=None)[source]#
Log loss, aka logistic loss or cross-entropy loss.
This is the loss function used in (multinomial) logistic regressionand extensions of it such as neural networks, defined as the negativelog-likelihood of a logistic model that returns
y_predprobabilitiesfor its training datay_true.The log loss is only defined for two or more labels.For a single sample with true label\(y \in \{0,1\}\) anda probability estimate\(p = \operatorname{Pr}(y = 1)\), the logloss is:\[L_{\log}(y, p) = -(y \log (p) + (1 - y) \log (1 - p))\]Read more in theUser Guide.
- Parameters:
- y_truearray-like or label indicator matrix
Ground truth (correct) labels for n_samples samples.
- y_predarray-like of float, shape = (n_samples, n_classes) or (n_samples,)
Predicted probabilities, as returned by a classifier’spredict_proba method. If
y_pred.shape=(n_samples,)the probabilities provided are assumed to be that of thepositive class. The labels iny_predare assumed to beordered alphabetically, as done byLabelBinarizer.y_predvalues are clipped to[eps,1-eps]whereepsis the machineprecision fory_pred’s dtype.- normalizebool, default=True
If true, return the mean loss per sample.Otherwise, return the sum of the per-sample losses.
- sample_weightarray-like of shape (n_samples,), default=None
Sample weights.
- labelsarray-like, default=None
If not provided, labels will be inferred from y_true. If
labelsisNoneandy_predhas shape (n_samples,) the labels areassumed to be binary and are inferred fromy_true.Added in version 0.18.
- Returns:
- lossfloat
Log loss, aka logistic loss or cross-entropy loss.
Notes
The logarithm used is the natural logarithm (base-e).
References
C.M. Bishop (2006). Pattern Recognition and Machine Learning. Springer,p. 209.
Examples
>>>fromsklearn.metricsimportlog_loss>>>log_loss(["spam","ham","ham","spam"],...[[.1,.9],[.9,.1],[.8,.2],[.35,.65]])0.21616
Gallery examples#
Probability Calibration for 3-class classification
Probabilistic predictions with Gaussian process classification (GPC)
