test

This is a standard PEFT LoRA derived from stabilityai/stable-diffusion-3-medium-diffusers.

The main validation prompt used during training was:

One canola seedling that is about 11 days old in a bright blue cylindrical cup on a bright blue background

Validation settings

CFG: 7.5
CFG Rescale: 0.0
Steps: 10
Sampler: None
Seed: 42
Resolution: 1024

Note: The validation settings are not necessarily the same as the training settings.

You can find some example images in the following gallery:

Prompt
unconditional (blank prompt)

Negative Prompt
blurry, floating leaves

Prompt
One canola seedling that is about 11 days old in a bright blue cylindrical cup on a bright blue background

Negative Prompt
blurry, floating leaves

The text encoder was not trained. You may reuse the base model text encoder for inference.

Training settings

Training epochs: 6
Training steps: 129
Learning rate: 0.00105
Max grad norm: 0.01
Effective batch size: 1
- Micro-batch size: 1
- Gradient accumulation steps: 1
- Number of GPUs: 1
Prediction type: flow-matching
Rescaled betas zero SNR: False
Optimizer: adamw_bf16
Precision: Pure BF16
Quantised: Yes: int8-quanto
Xformers: Not used
LoRA Rank: 256
LoRA Alpha: 256.0
LoRA Dropout: 0.1
LoRA initialisation style: default

Datasets

canola-test

Repeats: 1
Total number of images: 10
Total number of aspect buckets: 1
Resolution: 1024 px
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

Inference

import torch
from diffusers import DiffusionPipeline

model_id = 'stabilityai/stable-diffusion-3-medium-diffusers'
adapter_id = 'rdeinla/test'
pipeline = DiffusionPipeline.from_pretrained(model_id)
pipeline.load_lora_weights(adapter_id)

prompt = "One canola seedling that is about 11 days old in a bright blue cylindrical cup on a bright blue background"
negative_prompt = 'blurry, floating leaves'
pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu')
image = pipeline(
    prompt=prompt,
    negative_prompt=negative_prompt,
    num_inference_steps=10,
    generator=torch.Generator(device='cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu').manual_seed(1641421826),
    width=1024,
    height=1024,
    guidance_scale=7.5,
).images[0]
image.save("output.png", format="PNG")