Add datasets/fake/generate_dataset.py

datasets/vision/total_size.py

@@ -1,5 +1,13 @@
 import requests
 from lxml import html
+import os
+
+DOWNLOAD = True
+# An alternative for downloading would be something like:
+# ```bash
+# wget --reject 'index.html*' -l inf -nH --cut-dirs=2 --recursive --no-parent https://lesc.dinfo.unifi.it/VISION/dataset/
+# ```
+# but this would not only download `flat` images.
 
 url = 'https://lesc.dinfo.unifi.it/VISION/dataset/'
 
@@ -23,6 +31,8 @@ def getFolderEntries(url):
         entries += [[entryName, entrySize]]
     return entries
 
+os.mkdir('dataset')
+
 totalPhotoSizes = 0
 phoneFolders = getFolderEntries(url)
 for phoneFolder, _ in phoneFolders: