Skip to content

8627 perceptual loss errors out after hitting the maximum number of downloads #8652

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
virginiafdez wants to merge 16 commits into
base: dev
Choose a base branch
from
+52 −21
Open

8627 perceptual loss errors out after hitting the maximum number of downloads #8652

Changes from all commits
Commits
Show all changes
16 commits
Select commit Hold shift + click to select a range
b1e4a50
Perceptual loss changes.
Nov 26, 2025
0aeb4d9
Merge branch '8627-perceptual-loss-errors-out-after-hitting-the-maxim…
Nov 26, 2025
fa0639b
Fixes
Nov 26, 2025
685aee2
Merge branch 'dev' into 8627-perceptual-loss-errors-out-after-hitting…
virginiafdez Dec 2, 2025
915de5f
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Dec 2, 2025
5594bfe
Unnecessary import
Nov 26, 2025
b065de7
Merge branch '8627-perceptual-loss-errors-out-after-hitting-the-maxim…
Dec 5, 2025
c99e16e
Add check of network name
Dec 5, 2025
717b99b
Update monai/losses/perceptual.py
ericspod Dec 5, 2025
b276f3c
Update monai/losses/perceptual.py
ericspod Dec 5, 2025
2156b84
Update monai/losses/perceptual.py
ericspod Dec 5, 2025
e2b982e
Update monai/losses/perceptual.py
virginiafdez Dec 9, 2025
e3be8de
Bug
Dec 9, 2025
b02053b
Merge branch '8627-perceptual-loss-errors-out-after-hitting-the-maxim…
Dec 9, 2025
6dfc209
DCO Remediation Commit for Virginia Fernandez <virginia.fernandez@kcl…
Dec 9, 2025
d258390
Reformatting
Dec 9, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
73 changes: 52 additions & 21 deletions monai/losses/perceptual.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,11 +19,18 @@
from monai.utils import optional_import
from monai.utils.enums import StrEnum

# Valid model name to download from the repository
HF_MONAI_MODELS = frozenset(
("medicalnet_resnet10_23datasets", "medicalnet_resnet50_23datasets", "radimagenet_resnet50")
)
Comment on lines +22 to +25
Copy link
Contributor

@coderabbitai coderabbitai bot Dec 9, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟠 Major

Restrict model validation per family and guard 3D path to MedicalNet only.

As written, HF_MONAI_MODELS is shared by both MedicalNet and RadImageNet, and the 3D branch in PerceptualLoss always instantiates MedicalNetPerceptualSimilarity when spatial_dims == 3 and is_fake_3d is False, regardless of network_type. This leads to:

  • network_type="radimagenet_resnet50" with spatial_dims=3 and is_fake_3d=False being passed into MedicalNetPerceptualSimilarity, which will attempt to run a 2D RadImageNet backbone in a 3D MedicalNet path (shape/device errors at runtime instead of a clean validation failure).
  • MedicalNetPerceptualSimilarity and RadImageNetPerceptualSimilarity both accepting each other’s model names because they both use HF_MONAI_MODELS directly.

Recommend:

  1. Split the valid model-name set into MedicalNet-specific and RadImageNet-specific subsets and validate each class against its own subset.
  2. In the 3D branch, require that network_type is a MedicalNet variant before constructing MedicalNetPerceptualSimilarity, otherwise raise a clear ValueError.

Example patch:

@@
-# Valid model name to download from the repository
-HF_MONAI_MODELS = frozenset(
-    ("medicalnet_resnet10_23datasets", "medicalnet_resnet50_23datasets", "radimagenet_resnet50")
-)
+# Valid model names to download from the repository
+HF_MONAI_MODELS = frozenset(
+    ("medicalnet_resnet10_23datasets", "medicalnet_resnet50_23datasets", "radimagenet_resnet50")
+)
+HF_MONAI_MEDICALNET_MODELS = frozenset(
+    ("medicalnet_resnet10_23datasets", "medicalnet_resnet50_23datasets")
+)
+HF_MONAI_RADIMAGENET_MODELS = frozenset(("radimagenet_resnet50",))
@@
-        # If spatial_dims is 3, only MedicalNet supports 3D models, otherwise, spatial_dims=2 and fake_3D must be used.
-        if spatial_dims == 3 and is_fake_3d is False:
-            self.perceptual_function = MedicalNetPerceptualSimilarity(
-                net=network_type, verbose=False, channel_wise=channel_wise, cache_dir=cache_dir
-            )
+        # If spatial_dims is 3, only MedicalNet supports 3D models; other networks must use the fake 3D path.
+        if spatial_dims == 3 and is_fake_3d is False:
+            if "medicalnet_" not in network_type:
+                raise ValueError(
+                    "Only MedicalNet networks support 3D perceptual loss with is_fake_3d=False; "
+                    f"got network_type={network_type!r}."
+                )
+            self.perceptual_function = MedicalNetPerceptualSimilarity(
+                net=network_type, verbose=False, channel_wise=channel_wise, cache_dir=cache_dir
+            )
@@
-        if net not in HF_MONAI_MODELS:
-            raise ValueError(f"Invalid download model name '{net}'. Must be one of: {', '.join(HF_MONAI_MODELS)}.")
+        if net not in HF_MONAI_MEDICALNET_MODELS:
+            raise ValueError(
+                f"Invalid MedicalNet model name '{net}'. Must be one of: {', '.join(HF_MONAI_MEDICALNET_MODELS)}."
+            )
@@
-        if net not in HF_MONAI_MODELS:
-            raise ValueError(f"Invalid download model name '{net}'. Must be one of: {', '.join(HF_MONAI_MODELS)}.")
+        if net not in HF_MONAI_RADIMAGENET_MODELS:
+            raise ValueError(
+                f"Invalid RadImageNet model name '{net}'. Must be one of: {', '.join(HF_MONAI_RADIMAGENET_MODELS)}."
+            )

Also applies to: 94-107, 125-133, 234-239, 325-331


LPIPS, _ = optional_import("lpips", name="LPIPS")
torchvision, _ = optional_import("torchvision")


class PercetualNetworkType(StrEnum):
class PerceptualNetworkType(StrEnum):
"""Types of neural networks that are supported by perceptual loss."""

alex = "alex"
vgg = "vgg"
squeeze = "squeeze"
Expand Down Expand Up @@ -70,7 +77,7 @@ class PerceptualLoss(nn.Module):
def __init__(
self,
spatial_dims: int,
network_type: str = PercetualNetworkType.alex,
network_type: str = PerceptualNetworkType.alex,
is_fake_3d: bool = True,
fake_3d_ratio: float = 0.5,
cache_dir: str | None = None,
Expand All @@ -84,19 +91,25 @@ def __init__(
if spatial_dims not in [2, 3]:
raise NotImplementedError("Perceptual loss is implemented only in 2D and 3D.")

if (spatial_dims == 2 or is_fake_3d) and "medicalnet_" in network_type:
raise ValueError(
"MedicalNet networks are only compatible with ``spatial_dims=3``."
"Argument is_fake_3d must be set to False."
)

if channel_wise and "medicalnet_" not in network_type:
# Strict validation for MedicalNet
if "medicalnet_" in network_type:
if spatial_dims == 2 or is_fake_3d:
raise ValueError(
"MedicalNet networks are only compatible with ``spatial_dims=3``. Argument is_fake_3d must be set to False."
)
if not channel_wise:
warnings.warn(
"MedicalNet networks supp, ort channel-wise loss. Consider setting channel_wise=True.", stacklevel=2
)

# Channel-wise only for MedicalNet
elif channel_wise:
raise ValueError("Channel-wise loss is only compatible with MedicalNet networks.")

if network_type.lower() not in list(PercetualNetworkType):
if network_type.lower() not in list(PerceptualNetworkType):
raise ValueError(
"Unrecognised criterion entered for Adversarial Loss. Must be one in: %s"
% ", ".join(PercetualNetworkType)
% ", ".join(PerceptualNetworkType)
)
Comment on lines +109 to 113
Copy link
Contributor

@coderabbitai coderabbitai bot Dec 9, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟡 Minor

Error message still mentions “Adversarial Loss”.

The validation is for PerceptualLoss.network_type, but the message references “Adversarial Loss”, which is confusing for users.

-        if network_type.lower() not in list(PerceptualNetworkType):
-            raise ValueError(
-                "Unrecognised criterion entered for Adversarial Loss. Must be one in: %s"
-                % ", ".join(PerceptualNetworkType)
-            )
+        if network_type.lower() not in list(PerceptualNetworkType):
+            raise ValueError(
+                "Unrecognised network_type for PerceptualLoss. Must be one of: %s"
+                % ", ".join(PerceptualNetworkType)
+            )
🤖 Prompt for AI Agents 
In monai/losses/perceptual.py around lines 109-113, the ValueError message
incorrectly references "Adversarial Loss" while validating
PerceptualLoss.network_type; update the error text to mention "Perceptual Loss"
(or "PerceptualNetworkType") and list the valid network types (use the
appropriate enum names/values from PerceptualNetworkType) so the message reads
something like: "Unrecognised network_type for Perceptual Loss. Must be one of:
<allowed types>".


if cache_dir:
Expand All @@ -108,12 +121,16 @@ def __init__(

self.spatial_dims = spatial_dims
self.perceptual_function: nn.Module

# If spatial_dims is 3, only MedicalNet supports 3D models, otherwise, spatial_dims=2 and fake_3D must be used.
if spatial_dims == 3 and is_fake_3d is False:
self.perceptual_function = MedicalNetPerceptualSimilarity(
net=network_type, verbose=False, channel_wise=channel_wise
net=network_type, verbose=False, channel_wise=channel_wise, cache_dir=cache_dir
)
elif "radimagenet_" in network_type:
self.perceptual_function = RadImageNetPerceptualSimilarity(net=network_type, verbose=False)
self.perceptual_function = RadImageNetPerceptualSimilarity(
net=network_type, verbose=False, cache_dir=cache_dir
)
elif network_type == "resnet50":
self.perceptual_function = TorchvisionModelPerceptualSimilarity(
net=network_type,
Expand All @@ -122,7 +139,9 @@ def __init__(
pretrained_state_dict_key=pretrained_state_dict_key,
)
else:
# VGG, AlexNet and SqueezeNet are independently handled by LPIPS.
self.perceptual_function = LPIPS(pretrained=pretrained, net=network_type, verbose=False)

self.is_fake_3d = is_fake_3d
self.fake_3d_ratio = fake_3d_ratio
self.channel_wise = channel_wise
Expand Down Expand Up @@ -194,7 +213,7 @@ class MedicalNetPerceptualSimilarity(nn.Module):
"""
Component to perform the perceptual evaluation with the networks pretrained by Chen, et al. "Med3D: Transfer
Learning for 3D Medical Image Analysis". This class uses torch Hub to download the networks from
"Warvito/MedicalNet-models".
"Project-MONAI/perceptual-models".

Args:
net: {``"medicalnet_resnet10_23datasets"``, ``"medicalnet_resnet50_23datasets"``}
Expand All @@ -205,11 +224,19 @@ class MedicalNetPerceptualSimilarity(nn.Module):
"""

def __init__(
self, net: str = "medicalnet_resnet10_23datasets", verbose: bool = False, channel_wise: bool = False
self,
net: str = "medicalnet_resnet10_23datasets",
verbose: bool = False,
channel_wise: bool = False,
cache_dir: str | None = None,
) -> None:
super().__init__()
torch.hub._validate_not_a_forked_repo = lambda a, b, c: True
self.model = torch.hub.load("warvito/MedicalNet-models", model=net, verbose=verbose, trust_repo=True)
if net not in HF_MONAI_MODELS:
raise ValueError(f"Invalid download model name '{net}'. Must be one of: {', '.join(HF_MONAI_MODELS)}.")

self.model = torch.hub.load(
"Project-MONAI/perceptual-models:main", model=net, verbose=verbose, cache_dir=cache_dir, trust_repo=True
)
self.eval()

self.channel_wise = channel_wise
Expand Down Expand Up @@ -258,7 +285,7 @@ def forward(self, input: torch.Tensor, target: torch.Tensor) -> torch.Tensor:
for i in range(input.shape[1]):
l_idx = i * feats_per_ch
r_idx = (i + 1) * feats_per_ch
results[:, i, ...] = feats_diff[:, l_idx : i + r_idx, ...].sum(dim=1)
results[:, i, ...] = feats_diff[:, l_idx:r_idx, ...].sum(dim=1)
else:
results = feats_diff.sum(dim=1, keepdim=True)

Expand Down Expand Up @@ -287,17 +314,21 @@ class RadImageNetPerceptualSimilarity(nn.Module):
"""
Component to perform the perceptual evaluation with the networks pretrained on RadImagenet (pretrained by Mei, et
al. "RadImageNet: An Open Radiologic Deep Learning Research Dataset for Effective Transfer Learning"). This class
uses torch Hub to download the networks from "Warvito/radimagenet-models".
uses torch Hub to download the networks from "Project-MONAI/perceptual-models".

Args:
net: {``"radimagenet_resnet50"``}
Specifies the network architecture to use. Defaults to ``"radimagenet_resnet50"``.
verbose: if false, mute messages from torch Hub load function.
"""

def __init__(self, net: str = "radimagenet_resnet50", verbose: bool = False) -> None:
def __init__(self, net: str = "radimagenet_resnet50", verbose: bool = False, cache_dir: str | None = None) -> None:
super().__init__()
self.model = torch.hub.load("Warvito/radimagenet-models", model=net, verbose=verbose, trust_repo=True)
if net not in HF_MONAI_MODELS:
raise ValueError(f"Invalid download model name '{net}'. Must be one of: {', '.join(HF_MONAI_MODELS)}.")
self.model = torch.hub.load(
"Project-MONAI/perceptual-models:main", model=net, verbose=verbose, cache_dir=cache_dir, trust_repo=True
)
self.eval()

for param in self.parameters():
Expand Down
Loading