Text-to-image finetuning - Tochi2023/sd_martin_valen-model-v1-2_400_demo

This pipeline was finetuned from CompVis/stable-diffusion-v1-2 on the Ali-fb/martin_valen_dataset dataset. Below are some example images generated with the finetuned pipeline using the following prompts: ['black hoodie with a front half zipper by martin valen']:

Pipeline usage

You can use the pipeline like so:

from diffusers import DiffusionPipeline
import torch

pipeline = DiffusionPipeline.from_pretrained("Tochi2023/sd_martin_valen-model-v1-2_400_demo", torch_dtype=torch.float16)
prompt = "black hoodie with a front half zipper by martin valen"
image = pipeline(prompt).images[0]
image.save("my_image.png")

Training info

These are the key hyperparameters used during training:

• Epochs: 4
• Learning rate: 1e-05
• Batch size: 1
• Gradient accumulation steps: 4
• Image resolution: 512
• Mixed-precision: no