Instructions to use fofr/flux-handwriting with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use fofr/flux-handwriting with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("fofr/flux-handwriting") prompt = "HWRIT scrawling messy handwriting saying \"Hello, this is a handwriting lora\", illegible, red ink on old stained" image = pipe(prompt).images[0] - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
Flux-handwriting lora for low resource languages
#4
by CCRss - opened
Hello I'm trying to train Flux-handwriting lora for low resource language (kk) maybe you can help which parameters is best for training.
- Image:
- amount of images;
- text format;
- resolution;
- text length;
- Training params:
- learning rate;
- amount of epochs or steps;
- network_dim;
- network_alpha;
- other params that I might had missed;
Thanks in advance.π€