thehive/wide2D-datasets
Updated • 6
How to use thehive/wide2D with Diffusers:
pip install -U diffusers transformers accelerate
import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("thehive/wide2D", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]State of The Art model of Stable Diffusion version 1.5 that has been trained on 1k+ image from landscape anime/2D image..
The current model is finetuned with 2.0e-6 learning rate, 128.000 training steps.
Dataset has been preprocessed using Aspect Ratio Bucketing Tool.
I have utilized BLIP2 as a part of the training process as natural language prompts might be more effective.