Added warning message for dataset license

src/super_gradients/training/pretrained_models.py

@@ -61,7 +61,6 @@
     "yolo_nas_pose_l_coco_pose": "https://sghub.deci.ai/models/yolo_nas_pose_l_coco_pose.pth",
 }
 
-
 PRETRAINED_NUM_CLASSES = {
     "imagenet": 1000,
     "imagenet21k": 21843,
@@ -71,3 +70,13 @@
     "coco_pose": 17,
     "cifar10": 10,
 }
+
+DATASET_LICENSES = {
+    "imagenet": "https://www.image-net.org/download.php",
+    "imagenet21k": "https://github.com/Alibaba-MIIL/ImageNet21K",
+    "coco": "https://cocodataset.org/#termsofuse",
+    "coco_segmentation_subclass": "https://cocodataset.org/#termsofuse",
+    "coco_pose": "https://cocodataset.org/#termsofuse",
+    "cityscapes": "https://www.cs.toronto.edu/~kriz/cifar.html",
+    "objects365": "https://www.objects365.org/download.html",
+}

src/super_gradients/training/utils/checkpoint_utils.py

@@ -12,7 +12,7 @@
 from super_gradients.common.data_types import StrictLoad
 from super_gradients.common.decorators.explicit_params_validator import explicit_params_validation
 from super_gradients.module_interfaces import HasPredict
-from super_gradients.training.pretrained_models import MODEL_URLS
+from super_gradients.training.pretrained_models import MODEL_URLS, DATASET_LICENSES
 from super_gradients.training.utils.distributed_training_utils import wait_for_the_master
 from super_gradients.common.environment.ddp_utils import get_local_rank
 from super_gradients.training.utils.utils import unwrap_model
@@ -24,7 +24,6 @@
 except (ModuleNotFoundError, ImportError, NameError):
     from torch.hub import _download_url_to_file as download_url_to_file
 
-
 logger = get_logger(__name__)
 
 
@@ -52,8 +51,8 @@ def transfer_weights(model: nn.Module, model_state_dict: Mapping[str, Tensor]) -
     percentage_of_checkpoint = transfered_weights / len(model_state_dict)
     percentage_of_model = transfered_weights / len(model.state_dict())
     logger.debug(
-        f"Transfered {transfered_weights} ({(100*percentage_of_checkpoint):.2f}%) weights from the checkpoint. "
-        f"{(100*percentage_of_model):.2f}% of the model layers were initialized using checkpoint."
+        f"Transfered {transfered_weights} ({(100 * percentage_of_checkpoint):.2f}%) weights from the checkpoint. "
+        f"{(100 * percentage_of_model):.2f}% of the model layers were initialized using checkpoint."
     )
 
 
@@ -1562,6 +1561,13 @@ def load_pretrained_weights(model: torch.nn.Module, architecture: str, pretraine
     if model_url_key not in MODEL_URLS.keys():
         raise MissingPretrainedWeightsException(model_url_key)
 
+    if pretrained_weights in DATASET_LICENSES:
+        logger.warning(
+            f":warning: The pre-trained models provided by SuperGradients may have their own licenses or terms and "
+            "conditions derived from the dataset used for pre-training.\n It is your responsibility to determine whether you "
+            "have permission to use the models for your use case.\n The model you have requested was pre-trained on the "
+            f"{pretrained_weights} dataset, published under the following terms: {DATASET_LICENSES[pretrained_weights]}"
+        )
     url = MODEL_URLS[model_url_key]
 
     if architecture in {Models.YOLO_NAS_S, Models.YOLO_NAS_M, Models.YOLO_NAS_L}: