Bienqq
/
car-detection-bayes
2
1
Fork 0
Code Issues Pull Requests Releases Wiki Activity

P and R evaluated at 0.5 score

This commit is contained in:
Glenn Jocher 2020-03-07 10:26:08 -08:00
parent 65eeb1bae5
commit feea9c1a65
1 changed files with 4 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
utils/utils.py
View File

@ -188,6 +188,7 @@ def ap_per_class(tp, conf, pred_cls, target_cls):
unique_classes = np.unique(target_cls) unique_classes = np.unique(target_cls)
# Create Precision-Recall curve and compute AP for each class # Create Precision-Recall curve and compute AP for each class
pr_score = 0.5 # score to evaluate P and R https://github.com/ultralytics/yolov3/issues/898
s = [len(unique_classes), tp.shape[1]] # number class, number iou thresholds (i.e. 10 for mAP0.5...0.95) s = [len(unique_classes), tp.shape[1]] # number class, number iou thresholds (i.e. 10 for mAP0.5...0.95)
ap, p, r = np.zeros(s), np.zeros(s), np.zeros(s) ap, p, r = np.zeros(s), np.zeros(s), np.zeros(s)
for ci, c in enumerate(unique_classes): for ci, c in enumerate(unique_classes):
@ -204,18 +205,18 @@ def ap_per_class(tp, conf, pred_cls, target_cls):
# Recall # Recall
recall = tpc / (n_gt + 1e-16) # recall curve recall = tpc / (n_gt + 1e-16) # recall curve
r[ci] = recall[-1] r[ci] = np.interp(-pr_score, -conf[i], recall[:, 0]) # r at pr_score, negative x, xp because xp decreases
# Precision # Precision
precision = tpc / (tpc + fpc) # precision curve precision = tpc / (tpc + fpc) # precision curve
p[ci] = precision[-1] p[ci] = np.interp(-pr_score, -conf[i], precision[:, 0]) # p at pr_score
# AP from recall-precision curve # AP from recall-precision curve
for j in range(tp.shape[1]): for j in range(tp.shape[1]):
ap[ci, j] = compute_ap(recall[:, j], precision[:, j]) ap[ci, j] = compute_ap(recall[:, j], precision[:, j])
# Plot # Plot
# fig, ax = plt.subplots(1, 1, figsize=(4, 4)) # fig, ax = plt.subplots(1, 1, figsize=(5, 5))
# ax.plot(recall, precision) # ax.plot(recall, precision)
# ax.set_xlabel('Recall') # ax.set_xlabel('Recall')
# ax.set_ylabel('Precision') # ax.set_ylabel('Precision')