simpletuner-lora

This is a LyCORIS adapter derived from HiDream-ai/HiDream-I1-Full.

The main validation prompt used during training was:

1girl, cowboy shot

Validation settings

CFG: 2.5
CFG Rescale: 0.0
Steps: 50
Sampler: FlowMatchEulerDiscreteScheduler
Seed: 42
Resolution: 1024x1024

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, cropped, ugly

Prompt
1girl, cowboy shot

Negative Prompt
blurry, cropped, ugly

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

Training settings

Training epochs: 67
Training steps: 6000
Learning rate: 0.0001
- Learning rate schedule: constant_with_warmup
- Warmup steps: 100
Max grad value: 2.0
Effective batch size: 3
- Micro-batch size: 1
- Gradient accumulation steps: 1
- Number of GPUs: 3
Gradient checkpointing: True
Prediction type: flow_matching (extra parameters=['shift=3'])
Optimizer: adamw_bf16
Trainable parameter precision: Pure BF16
Base model precision: no_change
Caption dropout probability: 0.05%

LyCORIS Config:

{
    "algo": "lokr",
    "multiplier": 1.0,
    "linear_dim": 16384,
    "linear_alpha": 1,
    "full_matrix": true,
    "use_scalar": true,
    "factor": 16,
    "apply_preset": {
        "name_algo_map": {
            "double_stream_blocks.*.block.attn*": {
                "factor": 16
            },
            "double_stream_blocks.*.block.ff_t*": {
                "factor": 16
            },
            "double_stream_blocks.*.block.ff_i.shared_experts*": {
                "factor": 16
            },
            "single_stream_blocks.*.block.attn*": {
                "factor": 16
            },
            "single_stream_blocks.*.block.ff_i.shared_experts*": {
                "factor": 16
            }
        },
        "use_fnmatch": true
    }
}

Datasets

my-dataset-1024

Repeats: 0
Total number of images: ~267
Total number of aspect buckets: 1
Resolution: 1.048576 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

Inference

import torch
from diffusers import DiffusionPipeline
from lycoris import create_lycoris_from_weights


def download_adapter(repo_id: str):
    import os
    from huggingface_hub import hf_hub_download
    adapter_filename = "pytorch_lora_weights.safetensors"
    cache_dir = os.environ.get('HF_PATH', os.path.expanduser('~/.cache/huggingface/hub/models'))
    cleaned_adapter_path = repo_id.replace("/", "_").replace("\\", "_").replace(":", "_")
    path_to_adapter = os.path.join(cache_dir, cleaned_adapter_path)
    path_to_adapter_file = os.path.join(path_to_adapter, adapter_filename)
    os.makedirs(path_to_adapter, exist_ok=True)
    hf_hub_download(
        repo_id=repo_id, filename=adapter_filename, local_dir=path_to_adapter
    )

    return path_to_adapter_file
    
model_id = 'HiDream-ai/HiDream-I1-Full'
adapter_repo_id = 'Gohankaiju23/simpletuner-lora'
adapter_filename = 'pytorch_lora_weights.safetensors'
adapter_file_path = download_adapter(repo_id=adapter_repo_id)
pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16) # loading directly in bf16
lora_scale = 1.0
wrapper, _ = create_lycoris_from_weights(lora_scale, adapter_file_path, pipeline.transformer)
wrapper.merge_to()

prompt = "1girl, cowboy shot"
negative_prompt = 'blurry, cropped, ugly'

## Optional: quantise the model to save on vram.
## Note: The model was not quantised during training, so it is not necessary to quantise it during inference time.
#from optimum.quanto import quantize, freeze, qint8
#quantize(pipeline.transformer, weights=qint8)
#freeze(pipeline.transformer)
    
pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu') # the pipeline is already in its target precision level
model_output = pipeline(
    prompt=prompt,
    negative_prompt=negative_prompt,
    num_inference_steps=50,
    generator=torch.Generator(device='cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu').manual_seed(42),
    width=1024,
    height=1024,
    guidance_scale=2.5,
).images[0]

model_output.save("output.png", format="PNG")

Gohankaiju23
/

simpletuner-lora