metadata
license: other
base_model: terminusresearch/sana-1.6b-1024px
tags:
- sana
- sana-diffusers
- text-to-image
- diffusers
- simpletuner
- not-for-all-audiences
- lora
- template:sd-lora
- lycoris
inference: true
widget:
- text: unconditional (blank prompt)
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_0_0.png
- text: unconditional (blank prompt)
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_1_1.png
- text: A photo of mikrei eating a delicious slice of pizza at a restaurant
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_2_0.png
- text: A photo of mikrei eating a delicious slice of pizza at a restaurant
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_3_1.png
- text: >-
A photo of mikrei feeding a dinousaur in a jungle with lush vegetation and
cinematic lighting
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_4_0.png
- text: >-
A photo of mikrei feeding a dinousaur in a jungle with lush vegetation and
cinematic lighting
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_5_1.png
- text: >-
professional portrait of serious mikrei in a cockpit with instruments
piloting a luxurious private jet in dramatic weather
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_6_0.png
- text: >-
professional portrait of serious mikrei in a cockpit with instruments
piloting a luxurious private jet in dramatic weather
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_7_1.png
- text: >-
close-up portrait of mikrei on a throne in a magnificent palace wearing a
crown and royal attire
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_8_0.png
- text: >-
close-up portrait of mikrei on a throne in a magnificent palace wearing a
crown and royal attire
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_9_1.png
- text: >-
a high-quality, detailed photograph of mikrei as a sous-chef, immersed in
the art of culinary creation
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_10_0.png
- text: >-
a high-quality, detailed photograph of mikrei as a sous-chef, immersed in
the art of culinary creation
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_11_1.png
- text: >-
a lifelike and intimate portrait of mikrei, showcasing his unique
personality and charm
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_12_0.png
- text: >-
a lifelike and intimate portrait of mikrei, showcasing his unique
personality and charm
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_13_1.png
- text: >-
a cinematic, visually stunning photo of mikrei, emphasizing his dramatic
and captivating presence
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_14_0.png
- text: >-
a cinematic, visually stunning photo of mikrei, emphasizing his dramatic
and captivating presence
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_15_1.png
- text: >-
an elegant and timeless portrait of mikrei, exuding grace and
sophistication
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_16_0.png
- text: >-
an elegant and timeless portrait of mikrei, exuding grace and
sophistication
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_17_1.png
- text: >-
a dynamic and adventurous photo of mikrei, captured in an exciting,
action-filled moment
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_18_0.png
- text: >-
a dynamic and adventurous photo of mikrei, captured in an exciting,
action-filled moment
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_19_1.png
- text: >-
a mysterious and enigmatic portrait of mikrei, shrouded in shadows and
intrigue
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_20_0.png
- text: >-
a mysterious and enigmatic portrait of mikrei, shrouded in shadows and
intrigue
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_21_1.png
- text: >-
a vintage-style portrait of mikrei, evoking the charm and nostalgia of a
bygone era
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_22_0.png
- text: >-
a vintage-style portrait of mikrei, evoking the charm and nostalgia of a
bygone era
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_23_1.png
- text: >-
an artistic and abstract representation of mikrei, blending creativity
with visual storytelling
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_24_0.png
- text: >-
an artistic and abstract representation of mikrei, blending creativity
with visual storytelling
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_25_1.png
- text: >-
a futuristic and cutting-edge portrayal of mikrei, set against a backdrop
of advanced technology
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_26_0.png
- text: >-
a futuristic and cutting-edge portrayal of mikrei, set against a backdrop
of advanced technology
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_27_1.png
- text: >-
a beautifully crafted portrait of a woman, highlighting her natural beauty
and unique features
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_28_0.png
- text: >-
a beautifully crafted portrait of a woman, highlighting her natural beauty
and unique features
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_29_1.png
- text: >-
a powerful and striking portrait of a man, capturing his strength and
character
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_30_0.png
- text: >-
a powerful and striking portrait of a man, capturing his strength and
character
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_31_1.png
- text: >-
a playful and spirited portrait of a boy, capturing youthful energy and
innocence
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_32_0.png
- text: >-
a playful and spirited portrait of a boy, capturing youthful energy and
innocence
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_33_1.png
- text: >-
a charming and vibrant portrait of a girl, emphasizing her bright
personality and joy
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_34_0.png
- text: >-
a charming and vibrant portrait of a girl, emphasizing her bright
personality and joy
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_35_1.png
- text: >-
a heartwarming and cohesive family portrait, showcasing the bonds and
connections between loved ones
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_36_0.png
- text: >-
a heartwarming and cohesive family portrait, showcasing the bonds and
connections between loved ones
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_37_1.png
simpletuner-sana-lora-mikrei-1e-5
This is a LyCORIS adapter derived from terminusresearch/sana-1.6b-1024px.
No validation prompt was used during training.
None
Validation settings
- CFG:
4.0
- CFG Rescale:
0.0
- Steps:
30
- Sampler:
sana
- Seed:
42
- Resolutions:
1024x1024,1280x768
Note: The validation settings are not necessarily the same as the training settings.
You can find some example images in the following gallery:
The text encoder was not trained. You may reuse the base model text encoder for inference.
Training settings
- Training epochs: 0
- Training steps: 4500
- Learning rate: 1e-05
- Learning rate schedule: polynomial
- Warmup steps: 100
- Max grad norm: 2.0
- Effective batch size: 1
- Micro-batch size: 1
- Gradient accumulation steps: 1
- Number of GPUs: 1
- Gradient checkpointing: False
- Prediction type: epsilon (extra parameters=['training_scheduler_timestep_spacing=trailing', 'inference_scheduler_timestep_spacing=trailing'])
- Optimizer: optimi-stableadamw
- Trainable parameter precision: Pure BF16
- Caption dropout probability: 10.0%
LyCORIS Config:
{
"algo": "lokr",
"multiplier": 1.0,
"linear_dim": 10000,
"linear_alpha": 1,
"factor": 16,
"apply_preset": {
"target_module": [
"Attention",
"FeedForward"
],
"module_algo_map": {
"Attention": {
"factor": 16
},
"FeedForward": {
"factor": 8
}
}
}
}
Datasets
mikrei-data-512px
- Repeats: 1000
- Total number of images: 17
- Total number of aspect buckets: 1
- Resolution: 0.262144 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 = 'terminusresearch/sana-1.6b-1024px'
adapter_repo_id = 'mtreinik/simpletuner-sana-lora-mikrei-1e-5'
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 = "An astronaut is riding a horse through the jungles of Thailand."
negative_prompt = 'blurry, cropped, ugly'
## Optional: quantise the model to save on vram.
## Note: The model was quantised during training, and so it is recommended to do the same 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
image = pipeline(
prompt=prompt,
negative_prompt=negative_prompt,
num_inference_steps=30,
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=4.0,
guidance_rescale=0.0,
).images[0]
image.save("output.png", format="PNG")
Exponential Moving Average (EMA)
SimpleTuner generates a safetensors variant of the EMA weights and a pt file.
The safetensors file is intended to be used for inference, and the pt file is for continuing finetuning.
The EMA model may provide a more well-rounded result, but typically will feel undertrained compared to the full model as it is a running decayed average of the model weights.