Security News
tea.xyz Spam Plagues npm and RubyGems Package Registries
Tea.xyz, a crypto project aimed at rewarding open source contributions, is once again facing backlash due to an influx of spam packages flooding public package registries.
Readme
Deforum is a Python package for diffusion animation toolkit.
You can install Deforum using one of the following methods:
Install from PyPI using pip
:
pip install deforum
There's a provided bash script if you're on a Linux system.
./install-linux.sh
There's a provided batch file if you're on a Windows system.
install-windows.bat
Deforum has two sets of requirements which can be installed using:
pip install -r requirements.txt
For development requirements:
pip install -r requirements-dev.txt
Deforum is structured in following modules:
backend
: Contains the actual generation models. Options include base
for Stable Diffusion 1.5 and sdxl
for Stable Diffusion XL.
data
: Contains helper data for certain types of generation like wildcards, templates, prompts, stopwords, lightweight models.
modules
: Contains various helper classes and utilities for animation processing, controlnet auxiliary model processors, image transformation etc..
pipelines
: Contains pipeline classes which are used to generate images or videos using helper modules.
typed_classes
: Contains typed classes which are used for type validation and help.
utils
: Contains utilities for handling images, videos, and text outside of actual generation.
Here's a basic example:
import torch
from deforum import Deforum, DeforumConfig, GenerationArgs
config = DeforumConfig(
model_name="Lykon/AbsoluteReality",
model_type="sd1.5",
dtype=torch.float16,
)
deforum = Deforum(config)
args = GenerationArgs(
prompt="An ethereal cityscape under a starlit night sky",
negative_prompt="blurry, bright, devoid, and boring",
guidance_scale=7.5,
sampler="euler_ancestral",
num_inference_steps=30,
)
deforum.generate(args)
Deforum is licensed under the MIT License.
For more information please refer to license.
deforum/backend:
Actual model to be used for generation. Can be one of the following:
deforum/backend/mixins:
Superclasses which are used when instantiating classes from backend
deforum/backend/attn_processors:
Diffusers Attention processors
deforum/backend/model_loading:
Model loading factories for specific model types
deforum/backend/models:
Model classes for specific model types, which implement the actual __call__ function which performs the diffusion inference loop.
deforum/data:
Helper data for certain types of generation or misc things
deforum/modules:
Animation helper classes, controlnet auxiliary model processors for use DURING a generation...
deforum/pipelines:
Pipeline classes which are used to generate images or videos sample function implementation for the different models in backend, using helper modules from /modules
deforum/typed_classes:
Typed classes which are used for type validation and type help
deforum/utils:
Utilities for working with images / videos / text outside of actual generation
deforum/
backend/
mixins/
data/
modules/
pipelines/
typed_classes/
utils/
deforum.py
FAQs
diffusion animation toolkit
We found that deforum demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Security News
Tea.xyz, a crypto project aimed at rewarding open source contributions, is once again facing backlash due to an influx of spam packages flooding public package registries.
Security News
As cyber threats become more autonomous, AI-powered defenses are crucial for businesses to stay ahead of attackers who can exploit software vulnerabilities at scale.
Security News
UnitedHealth Group disclosed that the ransomware attack on Change Healthcare compromised protected health information for millions in the U.S., with estimated costs to the company expected to reach $1 billion.