# Module: tf.keras.metrics
|---| | [![](https://www.tensorflow.org/images/tf_logo_32px.png) TensorFlow 2 version](https://www.tensorflow.org/api_docs/python/tf/keras/metrics) | Built-in metrics. #### View aliases **Compat aliases for migration** See [Migration guide](https://www.tensorflow.org/guide/migrate) for more details. [`tf.compat.v1.keras.metrics`](https://www.tensorflow.org/api_docs/python/tf/compat/v1/keras/metrics)
## Classes [`class AUC`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/AUC): Computes the approximate AUC (Area under the curve) via a Riemann sum. [`class Accuracy`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/Accuracy): Calculates how often predictions matches labels. [`class BinaryAccuracy`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/BinaryAccuracy): Calculates how often predictions matches labels. [`class BinaryCrossentropy`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/BinaryCrossentropy): Computes the crossentropy metric between the labels and predictions. [`class CategoricalAccuracy`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/CategoricalAccuracy): Calculates how often predictions matches labels. [`class CategoricalCrossentropy`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/CategoricalCrossentropy): Computes the crossentropy metric between the labels and predictions. [`class CategoricalHinge`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/CategoricalHinge): Computes the categorical hinge metric between `y_true` and `y_pred`. [`class CosineSimilarity`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/CosineSimilarity): Computes the cosine similarity between the labels and predictions. [`class FalseNegatives`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/FalseNegatives): Calculates the number of false negatives. [`class FalsePositives`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/FalsePositives): Calculates the number of false positives. [`class Hinge`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/Hinge): Computes the hinge metric between `y_true` and `y_pred`. [`class KLDivergence`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/KLDivergence): Computes Kullback-Leibler divergence metric between `y_true` and `y_pred`. [`class LogCoshError`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/LogCoshError): Computes the logarithm of the hyperbolic cosine of the prediction error. [`class Mean`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/Mean): Computes the (weighted) mean of the given values. [`class MeanAbsoluteError`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/MeanAbsoluteError): Computes the mean absolute error between the labels and predictions. [`class MeanAbsolutePercentageError`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/MeanAbsolutePercentageError): Computes the mean absolute percentage error between `y_true` and `y_pred`. [`class MeanIoU`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/MeanIoU): Computes the mean Intersection-Over-Union metric. [`class MeanRelativeError`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/MeanRelativeError): Computes the mean relative error by normalizing with the given values. [`class MeanSquaredError`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/MeanSquaredError): Computes the mean squared error between `y_true` and `y_pred`. [`class MeanSquaredLogarithmicError`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/MeanSquaredLogarithmicError): Computes the mean squared logarithmic error between `y_true` and `y_pred`. [`class MeanTensor`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/MeanTensor): Computes the element-wise (weighted) mean of the given tensors. [`class Metric`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/Metric): Encapsulates metric logic and state. [`class Poisson`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/Poisson): Computes the Poisson metric between `y_true` and `y_pred`. [`class Precision`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/Precision): Computes the precision of the predictions with respect to the labels. [`class Recall`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/Recall): Computes the recall of the predictions with respect to the labels. [`class RootMeanSquaredError`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/RootMeanSquaredError): Computes root mean squared error metric between `y_true` and `y_pred`. [`class SensitivityAtSpecificity`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/SensitivityAtSpecificity): Computes the sensitivity at a given specificity. [`class SparseCategoricalAccuracy`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/SparseCategoricalAccuracy): Calculates how often predictions matches integer labels. [`class SparseCategoricalCrossentropy`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/SparseCategoricalCrossentropy): Computes the crossentropy metric between the labels and predictions. [`class SparseTopKCategoricalAccuracy`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/SparseTopKCategoricalAccuracy): Computes how often integer targets are in the top `K` predictions. [`class SpecificityAtSensitivity`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/SpecificityAtSensitivity): Computes the specificity at a given sensitivity. [`class SquaredHinge`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/SquaredHinge): Computes the squared hinge metric between `y_true` and `y_pred`. [`class Sum`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/Sum): Computes the (weighted) sum of the given values. [`class TopKCategoricalAccuracy`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/TopKCategoricalAccuracy): Computes how often targets are in the top `K` predictions. [`class TrueNegatives`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/TrueNegatives): Calculates the number of true negatives. [`class TruePositives`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/TruePositives): Calculates the number of true positives. ## Functions [`KLD(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/KLD): Computes Kullback-Leibler divergence loss between `y_true` and `y_pred`. [`MAE(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/MAE) [`MAPE(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/MAPE) [`MSE(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/MSE) [`MSLE(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/MSLE) [`binary_accuracy(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/binary_accuracy) [`binary_crossentropy(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/binary_crossentropy) [`categorical_accuracy(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/categorical_accuracy) [`categorical_crossentropy(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/categorical_crossentropy): Computes the categorical crossentropy loss. [`cosine(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/cosine): Computes the cosine similarity between labels and predictions. [`cosine_proximity(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/cosine): Computes the cosine similarity between labels and predictions. [`deserialize(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/deserialize) [`get(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/get) [`hinge(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/hinge): Computes the hinge loss between `y_true` and `y_pred`. [`kld(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/KLD): Computes Kullback-Leibler divergence loss between `y_true` and `y_pred`. [`kullback_leibler_divergence(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/KLD): Computes Kullback-Leibler divergence loss between `y_true` and `y_pred`. [`mae(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/MAE) [`mape(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/MAPE) [`mean_absolute_error(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/MAE) [`mean_absolute_percentage_error(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/MAPE) [`mean_squared_error(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/MSE) [`mean_squared_logarithmic_error(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/MSLE) [`mse(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/MSE) [`msle(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/MSLE) [`poisson(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/poisson): Computes the Poisson loss between y_true and y_pred. [`serialize(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/serialize) [`sparse_categorical_accuracy(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/sparse_categorical_accuracy) [`sparse_categorical_crossentropy(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/sparse_categorical_crossentropy) [`sparse_top_k_categorical_accuracy(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/sparse_top_k_categorical_accuracy) [`squared_hinge(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/losses/squared_hinge): Computes the squared hinge loss between `y_true` and `y_pred`. [`top_k_categorical_accuracy(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/keras/metrics/top_k_categorical_accuracy)