Skip to content

Enable local loading of tokenizer for dino.txt #531

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
alexcbb wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Enable local loading of tokenizer for dino.txt #531

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
13 changes: 9 additions & 4 deletions dinov2/hub/dinotxt.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -62,16 +62,21 @@ def dinov2_vitl14_reg4_dinotxt_tet1280d20h24l():
return model


def get_tokenizer():
def get_tokenizer(local_path=None):
Copy link
Contributor

@patricklabatut patricklabatut Jun 17, 2025

Choose a reason for hiding this comment

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

Minor: def get_tokenizer(bpe_path_or_url: Optional[str] = None)

from .text.tokenizer import Tokenizer
import requests
from io import BytesIO

url = _DINOV2_BASE_URL + "/thirdparty/bpe_simple_vocab_16e6.txt.gz"
try:
response = requests.get(url)
response.raise_for_status()
file_buf = BytesIO(response.content)
if not local_path:
Copy link
Contributor

@patricklabatut patricklabatut Jun 17, 2025

Choose a reason for hiding this comment

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

Use urllib.parse.urlparse(bpe_path).scheme to distinguish between a URL with a scheme or something that looks like an actual local path?

response = requests.get(url)
response.raise_for_status()
content = response.content
Comment on lines +73 to +75
Copy link
Contributor

@patricklabatut patricklabatut Jun 17, 2025

Choose a reason for hiding this comment

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

@cijose Do you remember why requests was pulled as dependency? Couldn't we simply use builtin Python modules instead?

    with urllib.request.urlopen(url) as f:
        content = f.read()

else:
with open(local_path, "rb") as f:
content = f.read()
file_buf = BytesIO(content)
return Tokenizer(vocab_path=file_buf)
except Exception as e:
raise FileNotFoundError(f"Failed to download file from url {url} with error last: {e}")