Fix: Vectorizing images and indexing on-demand

main.py

@@ -1,6 +1,6 @@
 #!/usr/bin/env python3
 from os import listdir, path, PathLike, remove
-from os.path import isfile, join
+from os.path import isfile, join, exists
 
 import pandas as pd
 import numpy as np
@@ -9,7 +9,7 @@ from PIL import ImageFile
 from tqdm import tqdm
 
 from uform import get_model
-from usearch.index import Index
+from usearch.index import Index, MetricKind
 from usearch.io import save_matrix, load_matrix
 
 ImageFile.LOAD_TRUNCATED_IMAGES = True
@@ -21,7 +21,7 @@ def is_image(path: PathLike) -> bool:
     try:
         Image.open(path)
         return True
-    except:
+    except Exception:
         return False
 
 
@@ -48,22 +48,25 @@ if len(names_to_delete) > 0:
     for name in names_to_delete:
         remove(join('images', name))
 
-model = get_model('unum-cloud/uform-vl-english')
-vectors = []
+if not exists('images.fbin'):
+    model = get_model('unum-cloud/uform-vl-english')
+    vectors = []
 
-for name in tqdm(names, desc='Vectorizing images'):
-    image = Image.open(join('images', name + '.jpg'))
-    image_data = model.preprocess_image(image)
-    image_embedding = model.encode_image(image_data).detach().numpy()
-    vectors.append(image_embedding)
+    for name in tqdm(names, desc='Vectorizing images'):
+        image = Image.open(join('images', name + '.jpg'))
+        image_data = model.preprocess_image(image)
+        image_embedding = model.encode_image(image_data).detach().numpy()
+        vectors.append(image_embedding)
 
-image_mat = np.concatenate(vectors)
-save_matrix(image_mat, 'images.fbin')
+    image_mat = np.concatenate(vectors)
+    save_matrix(image_mat, 'images.fbin')
 
-index = Index(ndim=256, metric='cos')
-image_mat = load_matrix('images.fbin')
+if not exists('images.usearch'):
+    index = Index(ndim=256, metric=MetricKind.Cos)
+    image_mat = load_matrix('images.fbin')
+    count = image_mat.shape[0]
 
-for idx, vector in tqdm(enumerate(vectors), desc='Indexing vectors'):
-    index.add(idx, vector.flatten())
+    for idx in tqdm(range(count), desc='Indexing vectors'):
+        index.add(idx, image_mat[idx, :].flatten())
 
-index.save('images.usearch')
+    index.save('images.usearch')