Upload 2 files

load_album.py

@@ -0,0 +1,45 @@
+# ----------------------------------------------------------------------------
+# Copyright (c) 2024 Amar Ali-bey
+#
+# OpenVPRLab: https://github.com/amaralibey/nanoCLIP
+#
+# Licensed under the MIT License. See LICENSE file in the project root.
+# ----------------------------------------------------------------------------
+
+from pathlib import Path
+from PIL import Image
+from torch.utils.data import Dataset
+from torchvision import transforms as T
+
+
+class AlbumDataset(Dataset):
+    def __init__(self, root_dir='./gallery/photos', transform=None):
+        """
+        This class is a simple dataset for loading ALL images from a directory and its subdirectories.
+        Formats supported: .jpg, .jpeg, .png, .bmp, .tiff
+        Args:
+            root_dir (str or Path): Path to the root directory containing images (e.g. gallery/).
+            transform (callable, optional): A function/transform to apply to the images.
+        """
+        self.root_dir = Path(root_dir)
+        if not self.root_dir.exists():
+            raise ValueError(f"Provided path {root_dir} does not exist.")
+        
+        # Gather all image paths
+        self.imgs = [p for p in self.root_dir.rglob('*') if p.suffix.lower() in ['.jpg', '.jpeg', '.png', '.bmp', '.tiff']]
+        if not self.imgs:
+            raise ValueError(f"No images found under {root_dir}.")
+        
+        self.transform = transform
+
+    def __len__(self):
+        return len(self.imgs)
+
+    def __getitem__(self, idx):
+        image_path = self.imgs[idx]
+        image = Image.open(image_path).convert("RGB")  # Ensure 3-channel RGB
+        
+        if self.transform:
+            image = self.transform(image)
+        
+        return image, str(image_path)  # Optionally return the path with the image

txt_encoder_state_dict.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f88cda86a69e423f48efb45dee12406a4d25f95d18a907ab112f9901503f630
+size 90994819