Crush Smol V0

标签

• 文本到视频
• diffusers训练
• diffusers
• cogvideox
• cogvideox - diffusers
• 模板：sd - lora

项目代码

项目代码可在 [GitHub](https://github.com/a - r - r - o - w/finetrainers) 上查看。

⚠️ 重要提示 这是一个实验性的检查点，其泛化能力较差是已知的情况。

🚀 快速开始

推理代码

以下是使用微调模型进行推理的代码示例：

from diffusers import CogVideoXTransformer3DModel, DiffusionPipeline 
from diffusers.utils import export_to_video
import torch 

transformer = CogVideoXTransformer3DModel.from_pretrained(
    "finetrainers/crush-smol-v0", torch_dtype=torch.bfloat16
)
pipeline = DiffusionPipeline.from_pretrained(
    "THUDM/CogVideoX-5b", transformer=transformer, torch_dtype=torch.bfloat16
).to("cuda")

prompt = """
DIFF_crush A thick burger is placed on a dining table, and a large metal cylinder descends from above, crushing the burger as if it were under a hydraulic press. The bulb is crushed, leaving a pile of debris around it.
"""
negative_prompt = "inconsistent motion, blurry motion, worse quality, degenerate outputs, deformed outputs"

video = pipeline(
    prompt=prompt, 
    negative_prompt=negative_prompt, 
    num_frames=81, 
    height=512,
    width=768,
    num_inference_steps=50
).frames[0]
export_to_video(video, "output.mp4", fps=25)

训练日志

训练日志可在WandB上查看：[点击查看](https://wandb.ai/sayakpaul/finetrainers - cogvideox/runs/ngcsyhom)

LoRA

我们从微调后的检查点中提取了一个秩为64的LoRA（提取脚本可查看 这里）。这个LoRA 可用于模拟相同的效果：

from diffusers import DiffusionPipeline 
from diffusers.utils import export_to_video
import torch 

pipeline = DiffusionPipeline.from_pretrained("THUDM/CogVideoX-5b", torch_dtype=torch.bfloat16).to("cuda")
pipeline.load_lora_weights("finetrainers/cakeify-v0", weight_name="extracted_crush_smol_lora_64.safetensors")

prompt = """
DIFF_crush A thick burger is placed on a dining table, and a large metal cylinder descends from above, crushing the burger as if it were under a hydraulic press. The bulb is crushed, leaving a pile of debris around it.
"""
negative_prompt = "inconsistent motion, blurry motion, worse quality, degenerate outputs, deformed outputs"

video = pipeline(
    prompt=prompt, 
    negative_prompt=negative_prompt, 
    num_frames=81, 
    height=512,
    width=768,
    num_inference_steps=50
).frames[0]
export_to_video(video, "output_lora.mp4", fps=25)