Skip to content

Conversation

@colbyford
Copy link
Contributor

This PR includes a Dockerfile, a simplified conda .yaml file, and an updated README.md that describes how to run FlowDock from inside the container.

I have pushed the Docker image to Docker Hub: https://hub.docker.com/repository/docker/cford38/flowdock

Notes:

  • We need to further simplify the environments/flowdock_environment_docker.yaml file to generalize the versions needed of each library. This current conda environment took hours to solve, likely due to the overly specific versions described in the .yaml file.
  • This has been tested on the inferencing functionality (python flowdock/sample.py ...) and inferences well with CUDA acceleration. However, it has not been tested on the training or evaluation FlowDock tasks.
  • This Docker image does not contain the checkpoints as it is already quite large (>12GB). Thus, users will need to download the desired checkpoint locally and run the Docker image with the local checkpoint directory mounted as a volume.

@colbyford colbyford mentioned this pull request Jan 24, 2025
@amorehead
Copy link
Collaborator

Thanks a ton for your contributions here, @colbyford!

@amorehead amorehead merged commit a66a9fd into BioinfoMachineLearning:main Jan 24, 2025
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