深度学习: mAP (Mean Average Precision)

版权声明：转载请注明出处 http://blog.csdn.net/JNingWei/article/details/78955536

mAP 概念

P

precision，即 准确率 。

R

recall，即 召回率 。

PR曲线

即 以 precision 和 recall 作为 横纵轴坐标 的二维曲线。

一般来说，precision 和 recall 是 鱼与熊掌 的关系。下图即是 PR曲线： 

AP值

Average Precision，即 平均精确度 。

如何衡量一个模型的性能，单纯用 precision 和 recall 都不科学。于是人们想到，哎嘛为何不把 PR曲线下的面积 当做衡量尺度呢？于是就有了 AP值 这一概念。这里的 average，等于是对 precision 进行 取平均 。

mAP值

Mean Average Precision，即 平均AP值 。

是对多个验证集个体 求 平均AP值 。如下图： 

mAP 计算

公式

Code

def compute_ap(gt_boxes, gt_class_ids,pred_boxes, pred_class_ids, pred_scores,iou_threshold=0.5):"""Compute Average Precision at a set IoU threshold (default 0.5).Returns:mAP: Mean Average Precisionprecisions: List of precisions at different class score thresholds.recalls: List of recall values at different class score thresholds.overlaps: [pred_boxes, gt_boxes] IoU overlaps."""# Trim zero padding and sort predictions by score from high to lowgt_boxes = trim_zeros(gt_boxes)pred_boxes = trim_zeros(pred_boxes)pred_scores = pred_scores[:pred_boxes.shape[0]]indices = np.argsort(pred_scores)[::-1]pred_boxes = pred_boxes[indices]pred_class_ids = pred_class_ids[indices]pred_scores = pred_scores[indices]# Compute IoU overlaps [pred_boxes, gt_boxes]overlaps = compute_overlaps(pred_boxes, gt_boxes)# Loop through ground truth boxes and find matching predictionsmatch_count = 0pred_match = np.zeros([pred_boxes.shape[0]])gt_match = np.zeros([gt_boxes.shape[0]])for i in range(len(pred_boxes)):# Find best matching ground truth boxsorted_ixs = np.argsort(overlaps[i])[::-1]for j in sorted_ixs:# If ground truth box is already matched, go to next oneif gt_match[j] == 1:continue# If we reach IoU smaller than the threshold, end the loopiou = overlaps[i, j]if iou < iou_threshold:break# Do we have a match?if pred_class_ids[i] == gt_class_ids[j]:match_count += 1gt_match[j] = 1pred_match[i] = 1break# Compute precision and recall at each prediction box stepprecisions = np.cumsum(pred_match) / (np.arange(len(pred_match)) + 1)recalls = np.cumsum(pred_match).astype(np.float32) / len(gt_match)# Pad with start and end values to simplify the mathprecisions = np.concatenate([[0], precisions, [0]])recalls = np.concatenate([[0], recalls, [1]])# Ensure precision values decrease but don't increase. This way, the# precision value at each recall threshold is the maximum it can be# for all following recall thresholds, as specified by the VOC paper.for i in range(len(precisions) - 2, -1, -1):precisions[i] = np.maximum(precisions[i], precisions[i + 1])# Compute mean AP over recall rangeindices = np.where(recalls[:-1] != recalls[1:])[0] + 1mAP = np.sum((recalls[indices] - recalls[indices - 1]) *precisions[indices])return mAP, precisions, recalls, overlaps