def validate(data_loader, net, loss, epoch, num_class):
start_time = time.time()
net.eval()
metrics = []
iou = 0
acc = 0
mean_acc = 0
for i, batch in enumerate(tqdm(data_loader)):
data, heatmaps, depth = batch['image'], batch['label'], batch['depth']
data = data.cuda(async=True)
#depth = depth.cuda(async=True)
heatmaps = heatmaps.cuda(async=True)
prediction = net(data)
loss_output = loss(prediction, heatmaps)
iou_, acc_, mean_acc_ = helper.cal_metric(heatmaps, prediction,
num_class)
iou += iou_
acc += acc_
mean_acc += mean_acc_
metrics.append(loss_output.item())
if i == (len(data_loader) - 2):
output_img = batch['image']
out_label = batch['label']
out_prediction = prediction
iou /= len(data_loader)
acc /= len(data_loader)
mean_acc /= len(data_loader)
img = helper.make_validation_img(output_img.numpy(), out_label.numpy(),
out_prediction.cpu().numpy())
cv2.imwrite('%s/validate_%d_%.4f.png' % (save_dir, epoch, iou),
img[:, :, ::-1])
end_time = time.time()
metrics = np.asarray(metrics, np.float32)
return metrics, end_time - start_time, iou, acc, mean_acc
def validate(data_loader, net, loss, epoch, num_class):
start_time = time.time()
net.eval()
metrics = []
iou = 0
acc = 0
mean_acc = 0
d_iou = 0
d_acc = 0
d_mean_acc = 0
k = 0
for i, batch in enumerate(tqdm(data_loader)):
data, heatmaps, depth = batch['image'], batch['label'], batch['depth']
data = data.cuda(async=True)
depth = depth.cuda(async=True)
heatmaps = heatmaps.cuda(async=True)
#prediction_d = net(data,depth)
prediction_d = net(data,torch.zeros(depth.size()))
#print(prediction_d.shape,prediction.shape)
#loss_output = loss(prediction, heatmaps)
#iou_, acc_, mean_acc_ = helper.cal_metric(heatmaps, prediction, num_class)
iou_d, acc_d, mean_acc_d = helper.cal_metric(heatmaps, prediction_d, num_class)
#iou += iou_
#acc += acc_
#mean_acc += mean_acc_
d_iou += iou_d
d_acc += acc_d
d_mean_acc += mean_acc_d
k += 1
if k==(len(data_loader)-2):
output_img = batch['image']
out_label = batch['label']
out_prediction_d = prediction_d
#out_prediction = prediction
#metrics.append(loss_output.item())
assert k==len(data_loader)
#iou /= len(data_loader)
#acc /= len(data_loader)
#mean_acc /= len(data_loader)
d_iou /= len(data_loader)
d_acc /= len(data_loader)
d_mean_acc /= len(data_loader)
#acc, mean_acc, iou, fwavacc = evaluate(predictions_all, gts_all, voc.num_classes)
img = helper.make_validation_img(output_img.numpy(),
out_label.numpy(),
out_prediction_d.cpu().numpy())
cv2.imwrite('%s/test_withdepth_%.4f.png'%(save_dir, iou),
img[:, :, ::-1])
end_time = time.time()
metrics = np.asarray(metrics, np.float32)
return metrics, end_time - start_time, d_iou, d_acc, d_mean_acc
def validate(data_loader, data_loader_1, data_loader_2, data_loader_3, net,
loss, epoch, num_class):
start_time = time.time()
net.eval()
metrics = []
iou = 0
acc = 0
mean_acc = 0
for batch, batch_1, batch_2, batch_3 in zip(data_loader, data_loader_1,
data_loader_2, data_loader_3):
data, heatmaps, depth = batch['image'], batch['label'], batch['depth']
data = data.cuda(async=True)
heatmaps_0 = heatmaps.cuda(async=True)
depth = depth.cuda(async=True)
prediction, loss_output, loss_1, loss_2, loss_3, loss_4 = net(
data, epoch, depth.squeeze(1), label=heatmaps, train=False)
prediction = F.interpolate(prediction,
scale_factor=1,
mode='bilinear',
align_corners=False).cpu()
data, heatmaps, depth = batch_1['image'], batch_1['label'], batch_1[
'depth']
data = data.cuda(async=True)
heatmaps = heatmaps.cuda(async=True)
depth = depth.cuda(async=True)
prediction_1, loss_output, loss_1, loss_2, loss_3, loss_4 = net(
data, epoch, depth.squeeze(1), label=heatmaps, train=False)
prediction_1 = F.interpolate(prediction_1,
scale_factor=5 / 12,
mode='bilinear',
align_corners=False).cpu()
data, heatmaps, depth = batch_2['image'], batch_2['label'], batch_2[
'depth']
data = data.cuda(async=True)
heatmaps = heatmaps.cuda(async=True)
depth = depth.cuda(async=True)
prediction_2, loss_output, loss_1, loss_2, loss_3, loss_4 = net(
data, epoch, depth.squeeze(1), label=heatmaps, train=False)
prediction_2 = F.interpolate(prediction_2,
scale_factor=1.25,
mode='bilinear',
align_corners=False).cpu()
data, heatmaps, depth = batch_3['image'], batch_3['label'], batch_3[
'depth']
data = data.cuda(async=True)
heatmaps = heatmaps.cuda(async=True)
depth = depth.cuda(async=True)
prediction_3, loss_output, loss_1, loss_2, loss_3, loss_4 = net(
data, epoch, depth.squeeze(1), label=heatmaps, train=False)
prediction_3 = F.interpolate(prediction_3,
scale_factor=0.625,
mode='bilinear',
align_corners=False).cpu()
pred = (prediction + prediction_1 + prediction_2 + prediction_3) / 4
iou_, acc_, mean_acc_ = helper.cal_metric(heatmaps_0.cpu(), pred,
num_class)
#iou_1, acc_1, mean_acc_1 = helper.cal_metric(label_list[i], pred_list_1[i], num_class)
#iou_2, acc_2, mean_acc_2 = helper.cal_metric(label_list[i], pred_list_2[i], num_class)
iou += iou_
#iou1 += iou_1
#iou2 += iou_2
acc += acc_
mean_acc += mean_acc_
iou /= len(data_loader)
#iou1 /= len(data_loader)
#iou2 /= len(data_loader)
acc /= len(data_loader)
mean_acc /= len(data_loader)
#img = helper.make_validation_img(batch['image'].numpy(),
# batch['label'].numpy(),
# prediction.cpu().numpy())
#cv2.imwrite('%s/validate_%d_%.4f.png'%(save_dir, epoch, iou),
# img[:, :, ::-1])
end_time = time.time()
return end_time - start_time, iou, acc, mean_acc
