Shekswess's picture
Update README.md
da61e8d verified
|
raw
history blame
4.14 kB
metadata
tags:
  - text-to-image
  - stable-diffusion
  - lora
  - diffusers
  - template:sd-lora
  - Neuron
  - Inferentia
widget:
  - text: >-
      Interior room of the house with plants, a chair and candles, space to
      relax, soft lighting, pastel colors, style of ultrafine detail, high
      quality photo --ar 2:3 --v 5
    output:
      url: images/11111.png
  - text: >-
      Interior of a van with blue water across the road, in the style of
      romantic landscape vistas, light beige and amber, sleepycore, maximalism,
      eco-friendly craftsmanship, luxurious drapery, photo taken with provia
      --ar 128:85 --v 6.0 --style raw
    output:
      url: images/22222.png
  - text: >-
      The modern living room by david carroll manne for stocksy united, in the
      style of highly detailed foliage, light brown, emerson silva, white
      background, leather/ hide, natural fibers, stock photo --v 5
    output:
      url: images/33333.png
base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: Interior Architecture, Interior Design, Modern Design
license: creativeml-openrail-m

Canopus-Interior-Architecture-0.1δ - Neuron

Prompt
Interior room of the house with plants, a chair and candles, space to relax, soft lighting, pastel colors, style of ultrafine detail, high quality photo --ar 2:3 --v 5
Prompt
Interior of a van with blue water across the road, in the style of romantic landscape vistas, light beige and amber, sleepycore, maximalism, eco-friendly craftsmanship, luxurious drapery, photo taken with provia --ar 128:85 --v 6.0 --style raw
Prompt
The modern living room by david carroll manne for stocksy united, in the style of highly detailed foliage, light brown, emerson silva, white background, leather/ hide, natural fibers, stock photo --v 5

Image Processing Parameters

Parameter Value Parameter Value
LR Scheduler constant Noise Offset 0.03
Optimizer AdamW Multires Noise Discount 0.1
Network Dim 64 Multires Noise Iterations 10
Network Alpha 32 Repeat & Steps 23 & 2200
Epoch 10 Save Every N Epochs 1

Trigger prompts

  Interior room of the house with plants, a chair and candles, space to relax, soft lighting, pastel colors, style of ultrafine detail, high quality photo --ar 2:3 --v 5

  Interior of a van with blue water across the road, in the style of romantic landscape vistas, light beige and amber, sleepycore, maximalism, eco-friendly craftsmanship, luxurious drapery, photo taken with provia --ar 128:85 --v 6.0 --style raw
  
  The modern living room by david carroll manne for stocksy united, in the style of highly detailed foliage, light brown, emerson silva, white background, leather/ hide, natural fibers, stock photo --v 5
Parameter Value
Prompt The modern living room by david carroll manne for stocksy united, in the style of highly detailed foliage, light brown, emerson silva, white background, leather/ hide, natural fibers, stock photo --v 5
Sampler euler

Trigger words

You should use Interior Architecture to trigger the image generation.

You should use Interior Design to trigger the image generation.

You should use Modern Design to trigger the image generation.

Usage

from diffusers import DPMSolverMultistepScheduler
from optimum.neuron import NeuronStableDiffusionXLInpaintPipeline

pipe = NeuronStableDiffusionXLInpaintPipeline.from_pretrained("Shekswess/Canopus-Interior-Architecture-0.1-Neuron", device_ids=[0, 1])
pipe.scheduler = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config)
prompt = "Interior of a van with blue water across the road, in the style of romantic landscape vistas, light beige and amber, sleepycore, maximalism, eco-friendly craftsmanship, luxurious drapery, photo taken with provia --ar 128:85 --v 6.0 --style raw"

image = pipe(
  prompt=prompt
).images[0].save("output.png")

Download model

Weights for this model are available in Safetensors format.

Download them in the Files & versions tab.

Original Model

Model

Precision

BFloat16 (bf16) For Matrix Multiplication Operations.