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

updates

This commit is contained in:
Glenn Jocher 2019-12-16 15:49:15 -08:00
parent 9064d42b93
commit d7b010c514
2 changed files with 29 additions and 31 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
test.py
View File

@ -13,7 +13,6 @@ def test(cfg,
weights=None,
batch_size=16,
img_size=416,
iou_thres=0.5,
conf_thres=0.001,
nms_thres=0.5,
save_json=False,
@ -49,6 +48,9 @@ def test(cfg,
nc = int(data['classes']) # number of classes
test_path = data['valid'] # path to test images
names = load_classes(data['names']) # class names
iou_thres = torch.linspace(0.5, 0.95, 10).to(device) # for mAP@0.5:0.95
iou_thres = iou_thres[0].view(1) # for mAP@0.5
niou = iou_thres.numel()
# Dataloader
if dataloader is None:
@ -120,7 +122,7 @@ def test(cfg,
clip_coords(pred, (height, width))
# Assign all predictions as incorrect
correct = [0] * len(pred)
correct = torch.zeros(len(pred), niou)
if nl:
detected = []
tcls_tensor = labels[:, 0]
@ -143,12 +145,13 @@ def test(cfg,
# Best iou, index between pred and targets
m = (pcls == tcls_tensor).nonzero().view(-1)
iou, bi = bbox_iou(pbox, tbox[m]).max(0)
iou, j = bbox_iou(pbox, tbox[m]).max(0)
m = m[j]
# If iou > threshold and class is correct mark as correct
if iou > iou_thres and m[bi] not in detected: # and pcls == tcls[bi]:
correct[i] = 1
detected.append(m[bi])
# Per iou_thres 'correct' vector
if iou > iou_thres[0] and m not in detected:
detected.append(m)
correct[i] = iou > iou_thres
# Append statistics (correct, conf, pcls, tcls)
stats.append((correct, pred[:, 4].cpu(), pred[:, 6].cpu(), tcls))
@ -157,6 +160,8 @@ def test(cfg,
stats = [np.concatenate(x, 0) for x in list(zip(*stats))] # to numpy
if len(stats):
p, r, ap, f1, ap_class = ap_per_class(*stats)
# if niou > 1:
# p, r, ap, f1 = p[:, 0], r[:, 0], ap[:, 0], ap.mean(1) # average across ious
mp, mr, map, mf1 = p.mean(), r.mean(), ap.mean(), f1.mean()
nt = np.bincount(stats[3].astype(np.int64), minlength=nc) # number of targets per class
else:
@ -208,7 +213,6 @@ if __name__ == '__main__':
parser.add_argument('--weights', type=str, default='weights/yolov3-spp.weights', help='path to weights file')
parser.add_argument('--batch-size', type=int, default=16, help='size of each image batch')
parser.add_argument('--img-size', type=int, default=416, help='inference size (pixels)')
parser.add_argument('--iou-thres', type=float, default=0.5, help='iou threshold required to qualify as detected')
parser.add_argument('--conf-thres', type=float, default=0.001, help='object confidence threshold')
parser.add_argument('--nms-thres', type=float, default=0.5, help='iou threshold for non-maximum suppression')
parser.add_argument('--save-json', action='store_true', help='save a cocoapi-compatible JSON results file')
@ -222,7 +226,6 @@ if __name__ == '__main__':
opt.weights,
opt.batch_size,
opt.img_size,
opt.iou_thres,
opt.conf_thres,
opt.nms_thres,
opt.save_json or any([x in opt.data for x in ['coco.data', 'coco2014.data', 'coco2017.data']]))

39
utils/utils.py
View File

@ -152,10 +152,10 @@ def ap_per_class(tp, conf, pred_cls, target_cls):
""" Compute the average precision, given the recall and precision curves.
Source: https://github.com/rafaelpadilla/Object-Detection-Metrics.
# Arguments
tp: True positives (list).
conf: Objectness value from 0-1 (list).
pred_cls: Predicted object classes (list).
target_cls: True object classes (list).
tp: True positives (nparray, nx1 or nx10).
conf: Objectness value from 0-1 (nparray).
pred_cls: Predicted object classes (nparray).
target_cls: True object classes (nparray).
# Returns
The average precision as computed in py-faster-rcnn.
"""
@ -168,46 +168,41 @@ def ap_per_class(tp, conf, pred_cls, target_cls):
unique_classes = np.unique(target_cls)
# Create Precision-Recall curve and compute AP for each class
ap, p, r = [], [], []
for c in unique_classes:
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)
for ci, c in enumerate(unique_classes):
i = pred_cls == c
n_gt = (target_cls == c).sum() # Number of ground truth objects
n_p = i.sum() # Number of predicted objects
n_gt = sum(target_cls == c) # Number of ground truth objects
n_p = sum(i) # Number of predicted objects
if n_p == 0 and n_gt == 0:
if n_p == 0 or n_gt == 0:
continue
elif n_p == 0 or n_gt == 0:
ap.append(0)
r.append(0)
p.append(0)
else:
# Accumulate FPs and TPs
fpc = (1 - tp[i]).cumsum()
tpc = (tp[i]).cumsum()
fpc = (1 - tp[i]).cumsum(0)
tpc = tp[i].cumsum(0)
# Recall
recall = tpc / (n_gt + 1e-16) # recall curve
r.append(recall[-1])
r[ci] = recall[-1]
# Precision
precision = tpc / (tpc + fpc) # precision curve
p.append(precision[-1])
p[ci] = precision[-1]
# AP from recall-precision curve
ap.append(compute_ap(recall, precision))
for j in range(tp.shape[1]):
ap[ci, j] = compute_ap(recall[:, j], precision[:, j])
# Plot
# fig, ax = plt.subplots(1, 1, figsize=(4, 4))
# ax.plot(np.concatenate(([0.], recall)), np.concatenate(([0.], precision)))
# ax.set_xlabel('YOLOv3-SPP')
# ax.set_xlabel('Recall')
# ax.set_ylabel('Precision')
# ax.set_title('YOLOv3-SPP'); ax.set_xlabel('Recall'); ax.set_ylabel('Precision')
# ax.set_xlim(0, 1)
# fig.tight_layout()
# fig.savefig('PR_curve.png', dpi=300)
# Compute F1 score (harmonic mean of precision and recall)
p, r, ap = np.array(p), np.array(r), np.array(ap)
f1 = 2 * p * r / (p + r + 1e-16)
return p, r, ap, f1, unique_classes.astype('int32')