Skip to content
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

v24.0.7 #2389

Merged
merged 25 commits into from
Apr 25, 2024
Merged

v24.0.7 #2389

merged 25 commits into from
Apr 25, 2024

Conversation

bmaltais
Copy link
Owner

  • Prevent crash if tkinter is not installed
  • Fix [24.0.6] Train toml config seed type error [24.0.6] Train toml config seed type error #2370
  • A new docker container is now built with every new release, eliminating the need for manual building. A big thank you to @jim60105 for his hard work in this area. You can find more information about it in the Docker section of the README.

bmaltais and others added 25 commits April 21, 2024 08:29
…ted-to-the-use_shell-variable

Fix issue with verify lora tool
- The Dockerfile no longer installs TensorFlow, only PyTorch
- The TensorBoard exposure port has been removed from the Dockerfile
- A new tensorboard service has been added to the docker-compose.yaml file, utilizing the TensorFlow:latest-gpu image and exposing port 6006.
- The requirements_linux_docker.txt file has the tensorboard and tensorflow requirements commented out.

Signed-off-by: 陳鈞 <[email protected]>
Remove tensorboard reduce the docker image size from 17GB to 10GB.
This reduction enables the image to be built on the GitHub free runner.

Signed-off-by: 陳鈞 <[email protected]>
- Modify the Docker publish workflow to only trigger on tag pushes that match the pattern `v*`, for instance, `v1.0.0`.
- Make the job run only on Ubuntu and tag commits.
- Enable submodules during checkout to pull all the data.
- Change Docker image name to `kohya-ss-gui`.
- Configure Docker metadata dynamically based on commit SHA and SemVer pattern.
- Set up QEMU, a generic and open source machine emulator and virtualizer, possibly for cross-platform docker image builds.
- Use `GITHUB_TOKEN` instead of `CR_PAT` for Docker login.
- Adjust the build arguments for Docker, including the version and the release number.
- Modify cache configuration to store in registry, rather than Github Action (gha) to avoid capacity limit.
- Enable software bill of materials (sbom) and provenance for better software supply chain and artifact traceability and verifiability.

Signed-off-by: 陳鈞 <[email protected]>
- Remove the mapping for the port 6006 in the docker-compose file

Signed-off-by: 陳鈞 <[email protected]>
…flow

- Removed the `github.sha` tag from the docker publish workflow file.

Signed-off-by: 陳鈞 <[email protected]>
- Add the `--do_not_use_shell` option to the Python CMD in the Dockerfile.

Signed-off-by: 陳鈞 <[email protected]>
This reverts commit 1bd0465.

`shell=False` does not work in Linux when passing the command as a string with arguments.
#2284 (comment)
…service

- Update the image source for `kohya-ss-gui` service in docker-compose file from `kohya-ss-gui:latest` to `ghcr.io/jim60105/kohya-ss-gui:latest`

`ghcr.io/jim60105/kohya-ss-gui` is for testing the forked repository and must be changed to `bmaltais`'s registry before merging the PR.

Signed-off-by: 陳鈞 <[email protected]>
…kage for writing logs.

- Removed exact version specification for tensorboard and tensorflow in the linux docker requirements, allowing the latest versions to be installed.
- Removed commented-out installations for tensorboard and tensorflow.

Signed-off-by: 陳鈞 <[email protected]>
- Added a new section in README.md about the design of the Dockerfile.
- Made changes in the Docker limitations section, including the separation of TensorBoard from the project and the related changes, no automatic browser launch, and manual file path input requirement.
- Updated instructions for docker upgrade using new code version.
- Note added for possible lengthy docker image build times.

Signed-off-by: 陳鈞 <[email protected]>
- Implemented rich logging in Dockerfile and set FORCE_COLOR environment variable to true
- Set the COLUMNS environment variable to `100` in Dockerfile for wider output
- Replaced single quotes with double quotes for `device_ids` value in docker-compose.yaml

Signed-off-by: 陳鈞 <[email protected]>
- Change the docker image from `ghcr.io/jim60105/kohya-ss-gui:latest` to `ghcr.io/bmaltais/kohya-ss-gui:latest`
- Update the cache source from `ghcr.io/jim60105/kohya-ss-gui:cache` to `ghcr.io/bmaltais/kohya-ss-gui:cache`

Signed-off-by: 陳鈞 <[email protected]>
…ed-type-error

Fix [24.0.6] Train toml config seed type error #2370
chore(docker): Decouple Tensorboard from the project and add CI build workflow.
@bmaltais bmaltais merged commit b826d61 into master Apr 25, 2024
4 checks passed
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