Deactivate CelebA download (#6052)

* Deactivate CelebA download

* flake8

* Do proto version

* Update torchvision/prototype/datasets/utils/_resource.py
Co-authored-by: Philip Meier <github.pmeier@posteo.de>

* address review
Co-authored-by: Philip Meier <github.pmeier@posteo.de>

torchvision/datasets/celeba.py

@@ -6,7 +6,7 @@ from typing import Any, Callable, List, Optional, Union, Tuple
 import PIL
 import torch
 
-from .utils import download_file_from_google_drive, check_integrity, verify_str_arg, extract_archive
+from .utils import check_integrity, verify_str_arg
 from .vision import VisionDataset
 
 CSV = namedtuple("CSV", ["header", "index", "data"])
@@ -35,9 +35,16 @@ class CelebA(VisionDataset):
             and returns a transformed version. E.g, ``transforms.PILToTensor``
         target_transform (callable, optional): A function/transform that takes in the
             target and transforms it.
-        download (bool, optional): If true, downloads the dataset from the internet and
-            puts it in root directory. If dataset is already downloaded, it is not
-            downloaded again.
+        download (bool, optional): Unsupported.
+
+            .. warning::
+
+                Downloading CelebA is not supported anymore as of 0.13. See
+                `this issue <https://github.com/pytorch/vision/issues/5705>`__
+                for more details.
+                Please download the files from
+                https://mmlab.ie.cuhk.edu.hk/projects/CelebA.html and extract
+                them in ``root/celeba``.
     """
 
     base_folder = "celeba"
@@ -146,10 +153,13 @@ class CelebA(VisionDataset):
             print("Files already downloaded and verified")
             return
 
-        for (file_id, md5, filename) in self.file_list:
-            download_file_from_google_drive(file_id, os.path.join(self.root, self.base_folder), filename, md5)
-
-        extract_archive(os.path.join(self.root, self.base_folder, "img_align_celeba.zip"))
+        raise ValueError(
+            "Downloading CelebA is not supported anymore as of 0.13. See "
+            "https://github.com/pytorch/vision/issues/5705 for more details. "
+            "Please download the files from "
+            "https://mmlab.ie.cuhk.edu.hk/projects/CelebA.html and extract them "
+            "in ``root/celeba``."
+        )
 
     def __getitem__(self, index: int) -> Tuple[Any, Any]:
         X = PIL.Image.open(os.path.join(self.root, self.base_folder, "img_align_celeba", self.filename[index]))

torchvision/prototype/datasets/_builtin/celeba.py

@@ -11,7 +11,7 @@ from torchdata.datapipes.iter import (
 )
 from torchvision.prototype.datasets.utils import (
     Dataset,
-    GDriveResource,
+    ManualDownloadResource,
     OnlineResource,
 )
 from torchvision.prototype.datasets.utils._internal import (
@@ -85,33 +85,34 @@ class CelebA(Dataset):
         super().__init__(root, skip_integrity_check=skip_integrity_check)
 
     def _resources(self) -> List[OnlineResource]:
-        splits = GDriveResource(
-            "0B7EVK8r0v71pY0NSMzRuSXJEVkk",
+        instructions = "Please download the file from https://mmlab.ie.cuhk.edu.hk/projects/CelebA.html."
+        splits = ManualDownloadResource(
+            instructions=instructions,
             sha256="fc955bcb3ef8fbdf7d5640d9a8693a8431b5f2ee291a5c1449a1549e7e073fe7",
             file_name="list_eval_partition.txt",
         )
-        images = GDriveResource(
-            "0B7EVK8r0v71pZjFTYXZWM3FlRnM",
+        images = ManualDownloadResource(
+            instructions=instructions,
             sha256="46fb89443c578308acf364d7d379fe1b9efb793042c0af734b6112e4fd3a8c74",
             file_name="img_align_celeba.zip",
         )
-        identities = GDriveResource(
-            "1_ee_0u7vcNLOfNLegJRHmolfH5ICW-XS",
+        identities = ManualDownloadResource(
+            instructions=instructions,
             sha256="c6143857c3e2630ac2da9f782e9c1232e5e59be993a9d44e8a7916c78a6158c0",
             file_name="identity_CelebA.txt",
         )
-        attributes = GDriveResource(
-            "0B7EVK8r0v71pblRyaVFSWGxPY0U",
+        attributes = ManualDownloadResource(
+            instructions=instructions,
             sha256="f0e5da289d5ccf75ffe8811132694922b60f2af59256ed362afa03fefba324d0",
             file_name="list_attr_celeba.txt",
         )
-        bounding_boxes = GDriveResource(
-            "0B7EVK8r0v71pbThiMVRxWXZ4dU0",
+        bounding_boxes = ManualDownloadResource(
+            instructions=instructions,
             sha256="7487a82e57c4bb956c5445ae2df4a91ffa717e903c5fa22874ede0820c8ec41b",
             file_name="list_bbox_celeba.txt",
         )
-        landmarks = GDriveResource(
-            "0B7EVK8r0v71pd0FJY3Blby1HUTQ",
+        landmarks = ManualDownloadResource(
+            instructions=instructions,
             sha256="6c02a87569907f6db2ba99019085697596730e8129f67a3d61659f198c48d43b",
             file_name="list_landmarks_align_celeba.txt",
         )

torchvision/prototype/datasets/utils/_resource.py

@@ -216,9 +216,9 @@ class ManualDownloadResource(OnlineResource):
 
     def _download(self, root: pathlib.Path) -> NoReturn:
         raise RuntimeError(
-            f"The file {self.file_name} cannot be downloaded automatically. "
-            f"Please follow the instructions below and place it in {root}\n\n"
-            f"{self.instructions}"
+            f"The file {self.file_name} was not found, and cannot be downloaded automatically.\n\n"
+            f"{self.instructions.strip()}\n\n"
+            f"Once it is downloaded, please place the file in {root}."
         )