def main():
model = Baseline(num_classes=702,
last_stride=1,
model_path=' ',
stn='no',
model_name='resnet50_ibn_a',
pretrain_choice=' ')
model.load_param('models/resnet50_ibn_a/duke_resnet50_ibn_a_model.pth')
model.to(device)
model.eval()
feats = []
with torch.no_grad():
img1 = process_img(
'/home/zzg/Datasets/DukeReiD/DukeMTMC-reID/query/0033_c1_f0057706.jpg'
)
feat1 = model(img1)
feats.append(feat1)
img2 = process_img(
'/home/zzg/Datasets/DukeReiD/DukeMTMC-reID/query/0033_c6_f0045755.jpg'
)
feat2 = model(img2)
feats.append(feat2)
img3 = process_img(
'/home/zzg/Datasets/DukeReiD/DukeMTMC-reID/query/0034_c2_f0057453.jpg'
)
feat3 = model(img3)
feats.append(feat3)
feats = torch.cat(feats, dim=0)
feats = torch.nn.functional.normalize(feats, dim=1, p=2)
dist = euclidean_dist_rank(feats, feats)
print(dist)
def main():
model = Baseline(num_classes=702,
last_stride=1,
model_path=' ',
stn='no',
model_name='resnet50_ibn_a',
pretrain_choice=' ')
model.load_param('models/resnet50_ibn_a/duke_resnet50_ibn_a_model.pth')
model.to(device)
model.eval()
feats = []
with torch.no_grad():
img1 = process_img(
'/nfs4/ajaym/Downloads/Ranked_Person_ReID-master/demo_data/1.jpg')
feat1 = model(img1)
feats.append(feat1)
img2 = process_img(
'/nfs4/ajaym/Downloads/Ranked_Person_ReID-master/demo_data/2.jpg')
feat2 = model(img2)
feats.append(feat2)
img3 = process_img(
'/nfs4/ajaym/Downloads/Ranked_Person_ReID-master/demo_data/3.jpg')
feat3 = model(img3)
feats.append(feat3)
feats = torch.cat(feats, dim=0)
feats = torch.nn.functional.normalize(feats, dim=1, p=2)
dist = euclidean_dist_rank(feats, feats)
print(dist)