tensorflow笔记（二十三）——tf.metrics

1.概述

tf.metrics主要是实现了评估相关的各种指标函数，比如mean、precesion、auc、precession、mse、precesion_at_k等等。以tf1.15为例，主要包括：

accuracy(...): Calculates how often predictions matches labels.
auc(...): Computes the approximate AUC via a Riemann sum.
average_precision_at_k(...): Computes average precision@k of predictions with respect to sparse labels.
false_negatives(...): Computes the total number of false negatives.
false_negatives_at_thresholds(...): Computes false negatives at provided threshold values.
false_positives(...): Sum the weights of false positives.
false_positives_at_thresholds(...): Computes false positives at provided threshold values.
mean(...): Computes the (weighted) mean of the given values.
mean_absolute_error(...): Computes the mean absolute error between the labels and predictions.
mean_cosine_distance(...): Computes the cosine distance between the labels and predictions.
mean_iou(...): Calculate per-step mean Intersection-Over-Union (mIOU).
mean_per_class_accuracy(...): Calculates the mean of the per-class accuracies.
mean_relative_error(...): Computes the mean relative error by normalizing with the given values.
mean_squared_error(...): Computes the mean squared error between the labels and predictions.
mean_tensor(...): Computes the element-wise (weighted) mean of the given tensors.
percentage_below(...): Computes the percentage of values less than the given threshold.
precision(...): Computes the precision of the predictions with respect to the labels.
precision_at_k(...): Computes precision@k of the predictions with respect to sparse labels.
precision_at_thresholds(...): Computes precision values for different thresholds on predictions.
precision_at_top_k(...): Computes precision@k of the predictions with respect to sparse labels.
recall(...)