valid argument fix for gradscaler

classify/val.py

@@ -108,7 +108,7 @@ def run(
     action = "validating" if dataloader.dataset.root.stem == "val" else "testing"
     desc = f"{pbar.desc[:-36]}{action:>36}" if pbar else f"{action}"
     bar = tqdm(dataloader, desc, n, not training, bar_format=TQDM_BAR_FORMAT, position=0)
-    with torch.amp.autocast('cuda', enabled=device.type != "cpu"):
+    with torch.amp.autocast("cuda", enabled=device.type != "cpu"):
         for images, labels in bar:
             with dt[0]:
                 images, labels = images.to(device, non_blocking=True), labels.to(device)

models/common.py

@@ -861,7 +861,7 @@ def forward(self, ims, size=640, augment=False, profile=False):
             p = next(self.model.parameters()) if self.pt else torch.empty(1, device=self.model.device)  # param
             autocast = self.amp and (p.device.type != "cpu")  # Automatic Mixed Precision (AMP) inference
             if isinstance(ims, torch.Tensor):  # torch
-                with torch.amp.autocast('cuda', enabled=autocast):
+                with torch.amp.autocast("cuda", enabled=autocast):
                     return self.model(ims.to(p.device).type_as(p), augment=augment)  # inference
 
             # Pre-process
@@ -888,7 +888,7 @@ def forward(self, ims, size=640, augment=False, profile=False):
             x = np.ascontiguousarray(np.array(x).transpose((0, 3, 1, 2)))  # stack and BHWC to BCHW
             x = torch.from_numpy(x).to(p.device).type_as(p) / 255  # uint8 to fp16/32
 
-        with torch.amp.autocast('cuda', enabled=autocast):
+        with torch.amp.autocast("cuda", enabled=autocast):
             # Inference
             with dt[1]:
                 y = self.model(x, augment=augment)  # forward

segment/train.py

@@ -320,7 +320,7 @@ def lf(x):
     maps = np.zeros(nc)  # mAP per class
     results = (0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)  # P, R, mAP@.5, mAP@.5-.95, val_loss(box, obj, cls)
     scheduler.last_epoch = start_epoch - 1  # do not move
-    scaler = torch.amp.GradScaler('cuda', enabled=amp)
+    scaler = torch.amp.GradScaler(enabled=amp)
     stopper, stop = EarlyStopping(patience=opt.patience), False
     compute_loss = ComputeLoss(model, overlap=overlap)  # init loss class
     # callbacks.run('on_train_start')
@@ -380,7 +380,7 @@ def lf(x):
                     imgs = nn.functional.interpolate(imgs, size=ns, mode="bilinear", align_corners=False)
 
             # Forward
-            with torch.amp.autocast('cuda', enabled=amp):
+            with torch.amp.autocast("cuda", enabled=amp):
                 pred = model(imgs)  # forward
                 loss, loss_items = compute_loss(pred, targets.to(device), masks=masks.to(device).float())
                 if RANK != -1:

train.py

@@ -409,7 +409,7 @@ def lf(x):
                     imgs = nn.functional.interpolate(imgs, size=ns, mode="bilinear", align_corners=False)
 
             # Forward
-            with torch.amp.autocast('cuda', enabled=amp):
+            with torch.amp.autocast("cuda", enabled=amp):
                 pred = model(imgs)  # forward
                 loss, loss_items = compute_loss(pred, targets.to(device))  # loss scaled by batch_size
                 if RANK != -1:

utils/autobatch.py

@@ -12,7 +12,7 @@
 
 def check_train_batch_size(model, imgsz=640, amp=True):
     """Checks and computes optimal training batch size for YOLOv5 model, given image size and AMP setting."""
-    with torch.amp.autocast('cuda', enabled=amp):
+    with torch.amp.autocast("cuda", enabled=amp):
         return autobatch(deepcopy(model).train(), imgsz)  # compute optimal batch size