File size: 1,009 Bytes
0e08f02 0b9d481 5246be9 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 | ---
language:
- en
tags:
- sdxl
- lora
- kaggle
- workflow
library_name: diffusers
pipeline_tag: text-to-image
license: mit
---
# mihai-lora-workflow (workflow-only)
This repository contains the reproducible training workflow used to run a personal SDXL LoRA pipeline on free Kaggle GPU, including checkpoint continuation, checkpoint evaluation, and LinkedIn-style gallery generation.
No personal training images, captions, generated portraits, or model checkpoints are included.
## Included
- Kaggle training script (`07_kaggle/train_flux_lora.py`)
- Config/run automation scripts (`03_configs/*`)
- Evaluation script templates (`08_kaggle_eval/*`)
- Runbook and process notes
## Excluded
- Raw/curated personal photos
- Captions tied to personal data
- Checkpoints and model weights
- Generated output galleries
- API tokens and credentials
## Privacy note
If you publish similar work, keep biometric data and personal LoRA weights private unless you explicitly want public distribution.
|