Instructions to use lambda/sd-pokemon-diffusers with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use lambda/sd-pokemon-diffusers with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("lambda/sd-pokemon-diffusers", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
model strangely overfit one image
#6
by Xiang-cd - opened
I tried to finetune the model with pokemon data at https://huggingface.co/datasets/lambdalabs/pokemon-blip-captions, but strangely that when sample or at val step of the train process. this butterfly is very often to appear, even with None prompt(un conditional generate), the example was here
My guess is that there were more images of it in the dataset than other Pokemon, since there are 18 different versions of Vivillon.