Closed fagemx closed 1 year ago
It already works :) check the examples folder
To be clear, this is what I'm using right now:
from stable_diffusion_videos import StableDiffusionWalkPipeline
from diffusers.models import AutoencoderKL
from diffusers.schedulers import LMSDiscreteScheduler
import torch
pipe = StableDiffusionWalkPipeline.from_pretrained(
'runwayml/stable-diffusion-v1-5',
vae=AutoencoderKL.from_pretrained(f"stabilityai/sd-vae-ft-ema"),
torch_dtype=torch.float16,
revision="fp16",
safety_checker=None,
scheduler=LMSDiscreteScheduler(
beta_start=0.00085, beta_end=0.012, beta_schedule="scaled_linear"
)
).to("cuda")
Closing this since it's not an issue.
Using the Stable-Diffusion-v1-5 checkpoint would be great!!!