Skip to content

LlamaFactory Playbook Review of PR #313#319

Merged
sreeram-11 merged 5 commits into
ldokovic/llama-factory-finetuningfrom
sreeram/review-313
Jun 1, 2026
Merged

LlamaFactory Playbook Review of PR #313#319
sreeram-11 merged 5 commits into
ldokovic/llama-factory-finetuningfrom
sreeram/review-313

Conversation

@sreeram-11
Copy link
Copy Markdown
Collaborator

@sreeram-11 sreeram-11 commented May 29, 2026

The below changes resolve the failing CI tests in PR #313.

  1. Increased the timeout for the quick-train-llamafactory-lora test to 3600s
    • The previous timeout caused failures on Strix and Kracken systems.
  2. Updated the Windows export-llamafactory-model test
    • Fixed incorrect Python referencing in the test command.
  3. Updated the Windows test code fence from bash to powershell
  4. Removed unnecessary <!-- @os:linux, windows --> tags
  5. Simplified the Windows LoRA fine-tuning configuration instructions
    • Previously, the instructions read the same YAML file, created the same backup, and wrote the file three separate times.
    • Updated the instructions to explain the three required YAML changes once and provide a single PowerShell block that updates all values in one pass.

Comment thread playbooks/supplemental/llama-factory-finetuning/README.md
@sreeram-11 sreeram-11 merged commit c1c70d7 into ldokovic/llama-factory-finetuning Jun 1, 2026
18 of 19 checks passed
ldokovic-personal added a commit that referenced this pull request Jun 2, 2026
…oss all OS's and GPU devices (#313)

* Playbook Improvements - making it stable for launch across all platforms and GPU devices

* LlamaFactory Playbook Review of PR #313 (#319)

* Cleanup

* Extended timeout for training, updated CI test for Windows training

* Updating Windows code fence to Powershell from Bash

* Updating export-llamafactory-model Windows test

* Keeping `@devices` tag to eliminate the need to create `~/.venv` on AI Halo devices

* Moving 'Create a Venv' title to avoid emtpty header on AIHalo

---------

Co-authored-by: Sreeram <s.sreeram.amd@gmail.com>
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