This pipeline was finetuned from sd2-community/stable-diffusion-2-1 on the ButterChicken98/CottonWeedID15_RAG_Captions dataset. Below are some example images generated with the finetuned pipeline using the following prompts: ['A close-up field photo of Morningglory weed, heart-shaped green leaves, natural soil background, daylight.', 'A close-up field photo of Carpetweed, small oval leaves in a low spreading cluster, natural soil background.', 'A close-up field photo of Palmer Amaranth weed seedling, pointed oval leaves, green upright stem, daylight.', 'A close-up field photo of Waterhemp weed plant, narrow lance-shaped leaves, green upright seedling, soil background.']:
val_imgs_grid
Pipeline usage
You can use the pipeline like so:
python
1from diffusers import DiffusionPipeline
2import torch
34pipeline = DiffusionPipeline.from_pretrained("ButterChicken98/sd21_cottonweed15_rag_k2_prompt_train", torch_dtype=torch.float16)5prompt ="A close-up field photo of Morningglory weed, heart-shaped green leaves, natural soil background, daylight."6image = pipeline(prompt).images[0]7image.save("my_image.png")
Training info
These are the key hyperparameters used during training:
Epochs: 25
Learning rate: 1e-05
Batch size: 8
Gradient accumulation steps: 1
Image resolution: 512
Mixed-precision: bf16
More information on all the CLI arguments and the environment are available on your wandb run page.
Intended uses & limitations
How to use
# TODO: add an example code snippet for running this diffusion pipeline
Limitations and bias
[TODO: provide examples of latent issues and potential remediations]
