Detection recipe enhancements (#5715)

* Detection recipe enhancements

* Add back nesterov momentum

Author: datumbox

references/classification/train.py

@@ -230,7 +230,7 @@ def main(args):
     criterion = nn.CrossEntropyLoss(label_smoothing=args.label_smoothing)
 
     if args.norm_weight_decay is None:
-        parameters = model.parameters()
+        parameters = [p for p in model.parameters() if p.requires_grad]
     else:
         param_groups = torchvision.ops._utils.split_normalization_params(model)
         wd_groups = [args.norm_weight_decay, args.weight_decay]

references/detection/presets.py

@@ -3,7 +3,7 @@
 
 
 class DetectionPresetTrain:
-    def __init__(self, data_augmentation, hflip_prob=0.5, mean=(123.0, 117.0, 104.0)):
+    def __init__(self, *, data_augmentation, hflip_prob=0.5, mean=(123.0, 117.0, 104.0)):
         if data_augmentation == "hflip":
             self.transforms = T.Compose(
                 [
@@ -12,6 +12,27 @@ def __init__(self, data_augmentation, hflip_prob=0.5, mean=(123.0, 117.0, 104.0)
                     T.ConvertImageDtype(torch.float),
                 ]
             )
+        elif data_augmentation == "lsj":
+            self.transforms = T.Compose(
+                [
+                    T.ScaleJitter(target_size=(1024, 1024)),
+                    T.FixedSizeCrop(size=(1024, 1024), fill=mean),
+                    T.RandomHorizontalFlip(p=hflip_prob),
+                    T.PILToTensor(),
+                    T.ConvertImageDtype(torch.float),
+                ]
+            )
+        elif data_augmentation == "multiscale":
+            self.transforms = T.Compose(
+                [
+                    T.RandomShortestSize(
+                        min_size=(480, 512, 544, 576, 608, 640, 672, 704, 736, 768, 800), max_size=1333
+                    ),
+                    T.RandomHorizontalFlip(p=hflip_prob),
+                    T.PILToTensor(),
+                    T.ConvertImageDtype(torch.float),
+                ]
+            )
         elif data_augmentation == "ssd":
             self.transforms = T.Compose(
                 [

references/detection/train.py

@@ -68,6 +68,7 @@ def get_args_parser(add_help=True):
     parser.add_argument(
         "-j", "--workers", default=4, type=int, metavar="N", help="number of data loading workers (default: 4)"
     )
+    parser.add_argument("--opt", default="sgd", type=str, help="optimizer")
     parser.add_argument(
         "--lr",
         default=0.02,
@@ -84,6 +85,12 @@ def get_args_parser(add_help=True):
         help="weight decay (default: 1e-4)",
         dest="weight_decay",
     )
+    parser.add_argument(
+        "--norm-weight-decay",
+        default=None,
+        type=float,
+        help="weight decay for Normalization layers (default: None, same value as --wd)",
+    )
     parser.add_argument(
         "--lr-scheduler", default="multisteplr", type=str, help="name of lr scheduler (default: multisteplr)"
     )
@@ -176,6 +183,8 @@ def main(args):
 
     print("Creating model")
     kwargs = {"trainable_backbone_layers": args.trainable_backbone_layers}
+    if args.data_augmentation in ["multiscale", "lsj"]:
+        kwargs["_skip_resize"] = True
     if "rcnn" in args.model:
         if args.rpn_score_thresh is not None:
             kwargs["rpn_score_thresh"] = args.rpn_score_thresh
@@ -191,8 +200,26 @@ def main(args):
         model = torch.nn.parallel.DistributedDataParallel(model, device_ids=[args.gpu])
         model_without_ddp = model.module
 
-    params = [p for p in model.parameters() if p.requires_grad]
-    optimizer = torch.optim.SGD(params, lr=args.lr, momentum=args.momentum, weight_decay=args.weight_decay)
+    if args.norm_weight_decay is None:
+        parameters = [p for p in model.parameters() if p.requires_grad]
+    else:
+        param_groups = torchvision.ops._utils.split_normalization_params(model)
+        wd_groups = [args.norm_weight_decay, args.weight_decay]
+        parameters = [{"params": p, "weight_decay": w} for p, w in zip(param_groups, wd_groups) if p]
+
+    opt_name = args.opt.lower()
+    if opt_name.startswith("sgd"):
+        optimizer = torch.optim.SGD(
+            parameters,
+            lr=args.lr,
+            momentum=args.momentum,
+            weight_decay=args.weight_decay,
+            nesterov="nesterov" in opt_name,
+        )
+    elif opt_name == "adamw":
+        optimizer = torch.optim.AdamW(parameters, lr=args.lr, weight_decay=args.weight_decay)
+    else:
+        raise RuntimeError(f"Invalid optimizer {args.opt}. Only SGD and AdamW are supported.")
 
     scaler = torch.cuda.amp.GradScaler() if args.amp else None
 

test/test_extended_models.py

@@ -64,7 +64,6 @@ def test_get_weight(name, weight):
 )
 def test_naming_conventions(model_fn):
     weights_enum = _get_model_weights(model_fn)
-    print(weights_enum)
     assert weights_enum is not None
     assert len(weights_enum) == 0 or hasattr(weights_enum, "DEFAULT")
 

torchvision/models/detection/faster_rcnn.py

@@ -187,6 +187,7 @@ def __init__(
         box_batch_size_per_image=512,
         box_positive_fraction=0.25,
         bbox_reg_weights=None,
+        **kwargs,
     ):
 
         if not hasattr(backbone, "out_channels"):
@@ -268,7 +269,7 @@ def __init__(
             image_mean = [0.485, 0.456, 0.406]
         if image_std is None:
             image_std = [0.229, 0.224, 0.225]
-        transform = GeneralizedRCNNTransform(min_size, max_size, image_mean, image_std)
+        transform = GeneralizedRCNNTransform(min_size, max_size, image_mean, image_std, **kwargs)
 
         super().__init__(backbone, rpn, roi_heads, transform)
 

torchvision/models/detection/fcos.py

@@ -373,6 +373,7 @@ def __init__(
         nms_thresh: float = 0.6,
         detections_per_img: int = 100,
         topk_candidates: int = 1000,
+        **kwargs,
     ):
         super().__init__()
         _log_api_usage_once(self)
@@ -410,7 +411,7 @@ def __init__(
             image_mean = [0.485, 0.456, 0.406]
         if image_std is None:
             image_std = [0.229, 0.224, 0.225]
-        self.transform = GeneralizedRCNNTransform(min_size, max_size, image_mean, image_std)
+        self.transform = GeneralizedRCNNTransform(min_size, max_size, image_mean, image_std, **kwargs)
 
         self.center_sampling_radius = center_sampling_radius
         self.score_thresh = score_thresh

torchvision/models/detection/keypoint_rcnn.py

@@ -198,6 +198,7 @@ def __init__(
         keypoint_head=None,
         keypoint_predictor=None,
         num_keypoints=None,
+        **kwargs,
     ):
 
         if not isinstance(keypoint_roi_pool, (MultiScaleRoIAlign, type(None))):
@@ -259,6 +260,7 @@ def __init__(
             box_batch_size_per_image,
             box_positive_fraction,
             bbox_reg_weights,
+            **kwargs,
         )
 
         self.roi_heads.keypoint_roi_pool = keypoint_roi_pool

torchvision/models/detection/mask_rcnn.py

@@ -195,6 +195,7 @@ def __init__(
         mask_roi_pool=None,
         mask_head=None,
         mask_predictor=None,
+        **kwargs,
     ):
 
         if not isinstance(mask_roi_pool, (MultiScaleRoIAlign, type(None))):
@@ -254,6 +255,7 @@ def __init__(
             box_batch_size_per_image,
             box_positive_fraction,
             bbox_reg_weights,
+            **kwargs,
         )
 
         self.roi_heads.mask_roi_pool = mask_roi_pool

torchvision/models/detection/retinanet.py

@@ -342,6 +342,7 @@ def __init__(
         fg_iou_thresh=0.5,
         bg_iou_thresh=0.4,
         topk_candidates=1000,
+        **kwargs,
     ):
         super().__init__()
         _log_api_usage_once(self)
@@ -383,7 +384,7 @@ def __init__(
             image_mean = [0.485, 0.456, 0.406]
         if image_std is None:
             image_std = [0.229, 0.224, 0.225]
-        self.transform = GeneralizedRCNNTransform(min_size, max_size, image_mean, image_std)
+        self.transform = GeneralizedRCNNTransform(min_size, max_size, image_mean, image_std, **kwargs)
 
         self.score_thresh = score_thresh
         self.nms_thresh = nms_thresh

torchvision/models/detection/ssd.py

@@ -195,6 +195,7 @@ def __init__(
         iou_thresh: float = 0.5,
         topk_candidates: int = 400,
         positive_fraction: float = 0.25,
+        **kwargs: Any,
     ):
         super().__init__()
         _log_api_usage_once(self)
@@ -227,7 +228,7 @@ def __init__(
         if image_std is None:
             image_std = [0.229, 0.224, 0.225]
         self.transform = GeneralizedRCNNTransform(
-            min(size), max(size), image_mean, image_std, size_divisible=1, fixed_size=size
+            min(size), max(size), image_mean, image_std, size_divisible=1, fixed_size=size, **kwargs
         )
 
         self.score_thresh = score_thresh

torchvision/models/detection/transform.py

@@ -1,5 +1,5 @@
 import math
-from typing import List, Tuple, Dict, Optional
+from typing import List, Tuple, Dict, Optional, Any
 
 import torch
 import torchvision
@@ -91,6 +91,7 @@ def __init__(
         image_std: List[float],
         size_divisible: int = 32,
         fixed_size: Optional[Tuple[int, int]] = None,
+        **kwargs: Any,
     ):
         super().__init__()
         if not isinstance(min_size, (list, tuple)):
@@ -101,6 +102,7 @@ def __init__(
         self.image_std = image_std
         self.size_divisible = size_divisible
         self.fixed_size = fixed_size
+        self._skip_resize = kwargs.pop("_skip_resize", False)
 
     def forward(
         self, images: List[Tensor], targets: Optional[List[Dict[str, Tensor]]] = None
@@ -170,6 +172,8 @@ def resize(
     ) -> Tuple[Tensor, Optional[Dict[str, Tensor]]]:
         h, w = image.shape[-2:]
         if self.training:
+            if self._skip_resize:
+                return image, target
             size = float(self.torch_choice(self.min_size))
         else:
             # FIXME assume for now that testing uses the largest scale

torchvision/ops/_utils.py

@@ -43,7 +43,13 @@ def split_normalization_params(
 ) -> Tuple[List[Tensor], List[Tensor]]:
     # Adapted from https://github.com/facebookresearch/ClassyVision/blob/659d7f78/classy_vision/generic/util.py#L501
     if not norm_classes:
-        norm_classes = [nn.modules.batchnorm._BatchNorm, nn.LayerNorm, nn.GroupNorm]
+        norm_classes = [
+            nn.modules.batchnorm._BatchNorm,
+            nn.LayerNorm,
+            nn.GroupNorm,
+            nn.modules.instancenorm._InstanceNorm,
+            nn.LocalResponseNorm,
+        ]
 
     for t in norm_classes:
         if not issubclass(t, nn.Module):