Skip to content

Install openvino-tokenizer python module in image #3274

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.

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
wants to merge 3 commits into
base: main
Choose a base branch
from

Conversation

stevegrubb
Copy link
Contributor

I was working through a demo where it said to run
"pip install -U optimum[diffusers,openvino]".
When I did that, I noticed that it downloaded:
openvino_tokenizers-2025.1.0.0-py3-none-manylinux2014_x86_64.whl. However, the tokenizer is built and installed in the image by the Dockerfile. After more digging, I found that openvino's python module is not installed in the image. This patch installs the openvino-tokenizer python module in the image and uses a symlink to point to the tokenizer shared object to save space.

stevegrubb added 2 commits May 5, 2025 15:32
I was working through a demo where it said to run
"pip install -U optimum[diffusers,openvino]".
When I did that, I noticed that it downloaded:
openvino_tokenizers-2025.1.0.0-py3-none-manylinux2014_x86_64.whl.
However, the tokenizer is built and installed in the image
by the Dockerfile. After more digging, I found that openvino's
python module is not installed in the image. This patch installs
the openvino-tokenizer python module in the image and uses a
symlink to point to the tokenizer shared object to save space.
Only create the convert_tokenizer if python is enbled
@mzegla
Copy link
Collaborator

mzegla commented May 19, 2025

@stevegrubb when working with which demo did you encounter this issue?

@stevegrubb
Copy link
Contributor Author

I think it was the stable_diffusion demo. I was logged in to an image built by Dockerfile.redhat. But I guess it doesn't matter which demo, because an end user can write code that uses the tokenizer and it should be preinstalled since we took the time to build it.

@stevegrubb
Copy link
Contributor Author

I updated the tokenizer version 2025.2.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants