metadata

license: mit
base_model: warp-ai/wuerstchen-prior
datasets:
  - haorandai/Mammal_Mice_lr0.01_e0.1_20_with20constraints
tags:
  - wuerstchen
  - text-to-image
  - diffusers
  - diffusers-training
inference: true

Finetuning - haorandai/temp

This pipeline was finetuned from warp-ai/wuerstchen-prior on the haorandai/Mammal_Mice_lr0.01_e0.1_20_with20constraints dataset. Below are some example images generated with the finetuned pipeline using the following prompts: ['An image of a mice and a cat']:

Pipeline usage

You can use the pipeline like so:

from diffusers import DiffusionPipeline
import torch

pipe_prior = DiffusionPipeline.from_pretrained("haorandai/temp", torch_dtype=torch.float16)
pipe_t2i = DiffusionPipeline.from_pretrained("warp-ai/wuerstchen", torch_dtype=torch.float16)
prompt = "An image of a mice and a cat"
(image_embeds,) = pipe_prior(prompt).to_tuple()
image = pipe_t2i(image_embeddings=image_embeds, prompt=prompt).images[0]
image.save("my_image.png")

Training info

These are the key hyperparameters used during training:

Epochs: 40
Learning rate: 1e-05
Batch size: 1
Gradient accumulation steps: 4
Image resolution: 512
Mixed-precision: fp16