def validate(loader, num_classes, device, net, criterion):
num_samples = 0
running_loss = 0
iou = MeanIoU(range(num_classes))
net.eval()
for images, masks, tiles in tqdm(loader, desc="Validate", unit="batch", ascii=True):
images = images.to(device)
masks = masks.to(device)
assert images.size()[2:] == masks.size()[1:], "resolutions for images and masks are in sync"
num_samples += int(images.size(0))
outputs = net(images)
assert outputs.size()[2:] == masks.size()[1:], "resolutions for predictions and masks are in sync"
assert outputs.size()[1] == num_classes, "classes for predictions and dataset are in sync"
loss = criterion(outputs, masks)
running_loss += loss.item()
for mask, output in zip(masks, outputs):
iou.add(mask.float(), output.max(0)[1].float())
assert num_samples > 0, "dataset contains validation images and labels"
return {"loss": running_loss / num_samples, "iou": iou.get()}
def validate(loader, num_classes, device, net, criterion):
num_samples = 0
running_loss = 0
iou = MeanIoU(range(num_classes))
net.eval()
for images, masks, tiles in tqdm(loader, desc='Validate', unit='batch', ascii=True):
images = images.to(device)
masks = masks.to(device)
assert images.size()[2:] == masks.size()[1:], 'resolutions for images and masks are in sync'
num_samples += int(images.size(0))
outputs = net(images)
assert outputs.size()[2:] == masks.size()[1:], 'resolutions for predictions and masks are in sync'
assert outputs.size()[1] == num_classes, 'classes for predictions and dataset are in sync'
loss = criterion(outputs, masks)
running_loss += loss.item()
for mask, output in zip(masks, outputs):
mask = mask.data.cpu().numpy()
prediction = output.data.max(0)[1].cpu().numpy()
iou.add(mask.ravel(), prediction.ravel())
assert num_samples > 0, 'dataset contains validation images and labels'
return {'loss': running_loss / num_samples, 'iou': iou.get()}
def validate(loader, num_classes, device, net, criterion):
num_samples = 0
running_loss = 0
iou = MeanIoU(range(num_classes))
net.eval()
for images, masks, tiles in tqdm(loader, desc='Validate', unit='batch', ascii=True):
images = images.to(device)
masks = masks.to(device)
assert images.size()[2:] == masks.size()[1:], 'resolutions for images and masks are in sync'
num_samples += int(images.size(0))
outputs = net(images)
assert outputs.size()[2:] == masks.size()[1:], 'resolutions for predictions and masks are in sync'
assert outputs.size()[1] == num_classes, 'classes for predictions and dataset are in sync'
loss = criterion(outputs, masks)
running_loss += loss.item()
for mask, output in zip(masks, outputs):
mask = mask.data.cpu().numpy()
prediction = output.data.max(0)[1].cpu().numpy()
iou.add(mask.ravel(), prediction.ravel())
return {'loss': running_loss / num_samples, 'iou': iou.get()}
def train(loader, num_classes, device, net, optimizer, scheduler, criterion):
num_samples = 0
running_loss = 0
iou = MeanIoU(range(num_classes))
net.train()
scheduler.step()
for images, masks, tiles in tqdm(loader,
desc="Train",
unit="batch",
ascii=True):
images = images.to(device)
masks = masks.to(device)
assert images.size()[2:] == masks.size(
)[1:], "resolutions for images and masks are in sync"
num_samples += int(images.size(0))
optimizer.zero_grad()
outputs = net(images)
assert outputs.size()[2:] == masks.size(
)[1:], "resolutions for predictions and masks are in sync"
assert outputs.size(
)[1] == num_classes, "classes for predictions and dataset are in sync"
loss = criterion(outputs, masks)
loss.backward()
optimizer.step()
running_loss += loss.item()
for mask, output in zip(masks, outputs):
mask = mask.data.cpu().numpy()
prediction = output.data.max(0)[1].cpu().numpy()
iou.add(mask.ravel(), prediction.ravel())
assert num_samples > 0, "dataset contains training images and labels"
return {"loss": running_loss / num_samples, "iou": iou.get()}