license: creativeml-openrail-m
base_model: amused/amused-512
datasets:
- lambdalabs/pokemon-blip-captions
tags:
- stable-diffusion
- stable-diffusion-diffusers
- text-to-image
- diffusers
- amused
inference: true
aMUSEd finetuning - suvadityamuk/amused-512-pokemon
This pipeline was finetuned from amused/amused-512 on the lambdalabs/pokemon-blip-captions dataset. Below are some example images generated with the finetuned pipeline using the following prompts: ['a pokemon red mammoth with unicorn horns', 'a pokemon blue fish with golden scales', 'a pokemon green goblin with glasses, wearing black pants and red shirt', 'a pokemon golden unicorn with shiny black hair and deep blue horns', 'a pokemon drawing of a dragon with its mouth closed', 'a pokemon red and yellow phoenix with fire on its wings', 'a pokemon purple tree with white leaves and golden nectar flowing', 'a pokemon green caterpillar']:
Pipeline usage
You can use the pipeline like so:
from diffusers import DiffusionPipeline
import torch
pipeline = DiffusionPipeline.from_pretrained("suvadityamuk/amused-512-pokemon", torch_dtype=torch.float16)
prompt = "a pokemon red mammoth with unicorn horns"
image = pipeline(prompt).images[0]
image.save("my_image.png")
Training info
These are the key hyperparameters used during training:
- Training Steps: 750
- Learning rate: 0.00025
- Batch size: 8
- Gradient accumulation steps: 4
- Image resolution: 512
- Mixed-precision: bf16
More information on all the CLI arguments and the environment are available on your wandb run page.