Merge branch 'main' into fix-make-image

Author: pmeier

docs/source/transforms.rst

@@ -234,7 +234,6 @@ Conversion
     v2.PILToTensor
     v2.ToImageTensor
     ConvertImageDtype
-    v2.ConvertDtype
     v2.ConvertImageDtype
     v2.ToDtype
     v2.ConvertBoundingBoxFormat

gallery/plot_transforms_v2_e2e.py

@@ -29,7 +29,7 @@ def show(sample):
     image, target = sample
     if isinstance(image, PIL.Image.Image):
         image = F.to_image_tensor(image)
-    image = F.convert_dtype(image, torch.uint8)
+    image = F.to_dtype(image, torch.uint8, scale=True)
     annotated_image = draw_bounding_boxes(image, target["boxes"], colors="yellow", width=3)
 
     fig, ax = plt.subplots()

references/detection/coco_utils.py

@@ -1,4 +1,3 @@
-import copy
 import os
 
 import torch
@@ -10,24 +9,6 @@
 from torchvision.datasets import wrap_dataset_for_transforms_v2
 
 
-class FilterAndRemapCocoCategories:
-    def __init__(self, categories, remap=True):
-        self.categories = categories
-        self.remap = remap
-
-    def __call__(self, image, target):
-        anno = target["annotations"]
-        anno = [obj for obj in anno if obj["category_id"] in self.categories]
-        if not self.remap:
-            target["annotations"] = anno
-            return image, target
-        anno = copy.deepcopy(anno)
-        for obj in anno:
-            obj["category_id"] = self.categories.index(obj["category_id"])
-        target["annotations"] = anno
-        return image, target
-
-
 def convert_coco_poly_to_mask(segmentations, height, width):
     masks = []
     for polygons in segmentations:
@@ -219,7 +200,7 @@ def __getitem__(self, idx):
         return img, target
 
 
-def get_coco(root, image_set, transforms, mode="instances", use_v2=False):
+def get_coco(root, image_set, transforms, mode="instances", use_v2=False, with_masks=False):
     anno_file_template = "{}_{}2017.json"
     PATHS = {
         "train": ("train2017", os.path.join("annotations", anno_file_template.format(mode, "train"))),
@@ -233,9 +214,12 @@ def get_coco(root, image_set, transforms, mode="instances", use_v2=False):
 
     if use_v2:
         dataset = torchvision.datasets.CocoDetection(img_folder, ann_file, transforms=transforms)
-        # TODO: need to update target_keys to handle masks for segmentation!
-        dataset = wrap_dataset_for_transforms_v2(dataset, target_keys={"boxes", "labels", "image_id"})
+        target_keys = ["boxes", "labels", "image_id"]
+        if with_masks:
+            target_keys += ["masks"]
+        dataset = wrap_dataset_for_transforms_v2(dataset, target_keys=target_keys)
     else:
+        # TODO: handle with_masks for V1?
         t = [ConvertCocoPolysToMask()]
         if transforms is not None:
             t.append(transforms)
@@ -249,9 +233,3 @@ def get_coco(root, image_set, transforms, mode="instances", use_v2=False):
     # dataset = torch.utils.data.Subset(dataset, [i for i in range(500)])
 
     return dataset
-
-
-def get_coco_kp(root, image_set, transforms, use_v2=False):
-    if use_v2:
-        raise ValueError("KeyPoints aren't supported by transforms V2 yet.")
-    return get_coco(root, image_set, transforms, mode="person_keypoints")

references/detection/train.py

@@ -28,7 +28,7 @@
 import torchvision.models.detection
 import torchvision.models.detection.mask_rcnn
 import utils
-from coco_utils import get_coco, get_coco_kp
+from coco_utils import get_coco
 from engine import evaluate, train_one_epoch
 from group_by_aspect_ratio import create_aspect_ratio_groups, GroupedBatchSampler
 from torchvision.transforms import InterpolationMode
@@ -42,10 +42,16 @@ def copypaste_collate_fn(batch):
 
 def get_dataset(is_train, args):
     image_set = "train" if is_train else "val"
-    paths = {"coco": (args.data_path, get_coco, 91), "coco_kp": (args.data_path, get_coco_kp, 2)}
-    p, ds_fn, num_classes = paths[args.dataset]
-
-    ds = ds_fn(p, image_set=image_set, transforms=get_transform(is_train, args), use_v2=args.use_v2)
+    num_classes, mode = {"coco": (91, "instances"), "coco_kp": (2, "person_keypoints")}[args.dataset]
+    with_masks = "mask" in args.model
+    ds = get_coco(
+        root=args.data_path,
+        image_set=image_set,
+        transforms=get_transform(is_train, args),
+        mode=mode,
+        use_v2=args.use_v2,
+        with_masks=with_masks,
+    )
     return ds, num_classes
 
 
@@ -68,7 +74,12 @@ def get_args_parser(add_help=True):
     parser = argparse.ArgumentParser(description="PyTorch Detection Training", add_help=add_help)
 
     parser.add_argument("--data-path", default="/datasets01/COCO/022719/", type=str, help="dataset path")
-    parser.add_argument("--dataset", default="coco", type=str, help="dataset name")
+    parser.add_argument(
+        "--dataset",
+        default="coco",
+        type=str,
+        help="dataset name. Use coco for object detection and instance segmentation and coco_kp for Keypoint detection",
+    )
     parser.add_argument("--model", default="maskrcnn_resnet50_fpn", type=str, help="model name")
     parser.add_argument("--device", default="cuda", type=str, help="device (Use cuda or cpu Default: cuda)")
     parser.add_argument(
@@ -171,6 +182,12 @@ def get_args_parser(add_help=True):
 def main(args):
     if args.backend.lower() == "datapoint" and not args.use_v2:
         raise ValueError("Use --use-v2 if you want to use the datapoint backend.")
+    if args.dataset not in ("coco", "coco_kp"):
+        raise ValueError(f"Dataset should be coco or coco_kp, got {args.dataset}")
+    if "keypoint" in args.model and args.dataset != "coco_kp":
+        raise ValueError("Oops, if you want Keypoint detection, set --dataset coco_kp")
+    if args.dataset == "coco_kp" and args.use_v2:
+        raise ValueError("KeyPoint detection doesn't support V2 transforms yet")
 
     if args.output_dir:
         utils.mkdir(args.output_dir)

test/common_utils.py

@@ -27,7 +27,7 @@
 from torch.testing._comparison import BooleanPair, NonePair, not_close_error_metas, NumberPair, TensorLikePair
 from torchvision import datapoints, io
 from torchvision.transforms._functional_tensor import _max_value as get_max_value
-from torchvision.transforms.v2.functional import convert_dtype_image_tensor, to_image_pil, to_image_tensor
+from torchvision.transforms.v2.functional import to_dtype_image_tensor, to_image_pil, to_image_tensor
 
 
 IN_OSS_CI = any(os.getenv(var) == "true" for var in ["CIRCLECI", "GITHUB_ACTIONS"])
@@ -602,7 +602,7 @@ def fn(shape, dtype, device, memory_format):
             image_tensor = image_tensor.to(device=device, memory_format=memory_format, copy=True)
         else:
             image_tensor = image_tensor.to(device=device)
-        image_tensor = convert_dtype_image_tensor(image_tensor, dtype=dtype)
+        image_tensor = to_dtype_image_tensor(image_tensor, dtype=dtype, scale=True)
 
         return datapoints.Image(image_tensor)
 

test/test_transforms_v2.py

@@ -1,7 +1,6 @@
 import itertools
 import pathlib
 import random
-import re
 import textwrap
 import warnings
 from collections import defaultdict
@@ -105,7 +104,7 @@ def normalize_adapter(transform, input, device):
             continue
         elif check_type(value, (datapoints.Image, datapoints.Video, is_simple_tensor)):
             # normalize doesn't support integer images
-            value = F.convert_dtype(value, torch.float32)
+            value = F.to_dtype(value, torch.float32, scale=True)
         adapted_input[key] = value
     return adapted_input
 
@@ -146,7 +145,7 @@ class TestSmoke:
             (transforms.ScaleJitter((16, 16), scale_range=(0.8, 1.2), antialias=True), None),
             (transforms.ClampBoundingBox(), None),
             (transforms.ConvertBoundingBoxFormat(datapoints.BoundingBoxFormat.CXCYWH), None),
-            (transforms.ConvertDtype(), None),
+            (transforms.ConvertImageDtype(), None),
             (transforms.GaussianBlur(kernel_size=3), None),
             (
                 transforms.LinearTransformation(
@@ -1326,61 +1325,6 @@ def test__transform(self, mocker):
         )
 
 
-class TestToDtype:
-    @pytest.mark.parametrize(
-        ("dtype", "expected_dtypes"),
-        [
-            (
-                torch.float64,
-                {
-                    datapoints.Video: torch.float64,
-                    datapoints.Image: torch.float64,
-                    datapoints.BoundingBox: torch.float64,
-                },
-            ),
-            (
-                {datapoints.Video: torch.int32, datapoints.Image: torch.float32, datapoints.BoundingBox: torch.float64},
-                {datapoints.Video: torch.int32, datapoints.Image: torch.float32, datapoints.BoundingBox: torch.float64},
-            ),
-        ],
-    )
-    def test_call(self, dtype, expected_dtypes):
-        sample = dict(
-            video=make_video(dtype=torch.int64),
-            image=make_image(dtype=torch.uint8),
-            bounding_box=make_bounding_box(format=datapoints.BoundingBoxFormat.XYXY, dtype=torch.float32),
-            str="str",
-            int=0,
-        )
-
-        transform = transforms.ToDtype(dtype)
-        transformed_sample = transform(sample)
-
-        for key, value in sample.items():
-            value_type = type(value)
-            transformed_value = transformed_sample[key]
-
-            # make sure the transformation retains the type
-            assert isinstance(transformed_value, value_type)
-
-            if isinstance(value, torch.Tensor):
-                assert transformed_value.dtype is expected_dtypes[value_type]
-            else:
-                assert transformed_value is value
-
-    @pytest.mark.filterwarnings("error")
-    def test_plain_tensor_call(self):
-        tensor = torch.empty((), dtype=torch.float32)
-        transform = transforms.ToDtype({torch.Tensor: torch.float64})
-
-        assert transform(tensor).dtype is torch.float64
-
-    @pytest.mark.parametrize("other_type", [datapoints.Image, datapoints.Video])
-    def test_plain_tensor_warning(self, other_type):
-        with pytest.warns(UserWarning, match=re.escape("`torch.Tensor` will *not* be transformed")):
-            transforms.ToDtype(dtype={torch.Tensor: torch.float32, other_type: torch.float64})
-
-
 class TestUniformTemporalSubsample:
     @pytest.mark.parametrize(
         "inpt",

test/test_transforms_v2_consistency.py

@@ -191,7 +191,7 @@ def __init__(
         closeness_kwargs=dict(rtol=None, atol=None),
     ),
     ConsistencyConfig(
-        v2_transforms.ConvertDtype,
+        v2_transforms.ConvertImageDtype,
         legacy_transforms.ConvertImageDtype,
         [
             ArgsKwargs(torch.float16),

test/test_transforms_v2_functional.py

@@ -283,12 +283,12 @@ def test_float32_vs_uint8(self, test_id, info, args_kwargs):
         adapted_other_args, adapted_kwargs = info.float32_vs_uint8(other_args, kwargs)
 
         actual = info.kernel(
-            F.convert_dtype_image_tensor(input, dtype=torch.float32),
+            F.to_dtype_image_tensor(input, dtype=torch.float32, scale=True),
             *adapted_other_args,
             **adapted_kwargs,
         )
 
-        expected = F.convert_dtype_image_tensor(info.kernel(input, *other_args, **kwargs), dtype=torch.float32)
+        expected = F.to_dtype_image_tensor(info.kernel(input, *other_args, **kwargs), dtype=torch.float32, scale=True)
 
         assert_close(
             actual,
@@ -538,7 +538,6 @@ def test_bounding_box_format_consistency(self, info, args_kwargs):
             (F.get_image_num_channels, F.get_num_channels),
             (F.to_pil_image, F.to_image_pil),
             (F.elastic_transform, F.elastic),
-            (F.convert_image_dtype, F.convert_dtype_image_tensor),
             (F.to_grayscale, F.rgb_to_grayscale),
         ]
     ],
@@ -547,24 +546,6 @@ def test_alias(alias, target):
     assert alias is target
 
 
-@pytest.mark.parametrize(
-    ("info", "args_kwargs"),
-    make_info_args_kwargs_params(
-        KERNEL_INFOS_MAP[F.convert_dtype_image_tensor],
-        args_kwargs_fn=lambda info: info.sample_inputs_fn(),
-    ),
-)
-@pytest.mark.parametrize("device", cpu_and_cuda())
-def test_convert_dtype_image_tensor_dtype_and_device(info, args_kwargs, device):
-    (input, *other_args), kwargs = args_kwargs.load(device)
-    dtype = other_args[0] if other_args else kwargs.get("dtype", torch.float32)
-
-    output = info.kernel(input, dtype)
-
-    assert output.dtype == dtype
-    assert output.device == input.device
-
-
 @pytest.mark.parametrize("device", cpu_and_cuda())
 @pytest.mark.parametrize("num_channels", [1, 3])
 def test_normalize_image_tensor_stats(device, num_channels):