WebGL 기반 3D 플롯, NumPy 배열에서 메쉬·등고선·볼륨·복셀·포인트 클라우드 시각화와 경로 추적 렌더러 제공
WebGL 3D plotting for Jupyter: meshes, isosurfaces, volumes, voxels and point clouds from NumPy arrays, with a path-traced cinematic renderer.
데이터Python · JavaScriptGitHub @K3D-tools직접 설치운영자 미확인
써 보셨어요? 한 번만 눌러 주세요. 로그인은 필요 없어요.
최근 완료된 8시간 회차 사이 GitHub 스타 증가량 순
FactCircuit는 불변 출처와 단계별 검증, 제한된 검색, 재현 가능한 평가를 제공하는 감사 가능한 주장·증거 검증 루프입니다
AI 에이전트를 위한 경량 Postgres, Rust로 구현, 하나의 런타임에서 수천 개의 격리된 데이터베이스 제공
그래프 엔진으로 컨텍스트를 조립하고, 객체 저장소 기반에서 에이전트가 코드처럼 그래프 데이터를 분기·타입·검토·병합·감사·거버넌스합니다
SeekStorm – 러스트 기반 벡터·전체 텍스트 검색, 임베드 라이브러리·자체 서버·클라우드 서비스 중 선택 가능
대규모 언어 모델을 위한 확장 가능한 데이터 전처리·정제 툴킷
메이커가 쓴 소개는 위의 한 줄이 전부입니다. 더 자세한 내용은 저장소 README에서 읽을 수 있습니다. 저장소 README ↗
README 전체를 불러오는 중입니다.
README 전문 보기 ↗K3D lets you create 3D plots backed by WebGL with high-level API (surfaces, isosurfaces, voxels, mesh, point clouds, vtk objects, volume renderer, colormaps, etc). The primary aim of K3D-jupyter is to be easy to use as a standalone package like matplotlib, but also to allow interoperation with existing libraries as VTK. K3D can be run as:
Since 3.0.0 a plot also chooses how it is lit, through plot.renderer: simple (the
default rasteriser), advanced (image-based lighting with ambient occlusion) or cinematic
(progressive path tracing). cinematic is experimental — it needs WebGL2 with renderable
float textures, does not cover every object (volume_slice is not drawn, mip stays
outside the light simulation, only the first volume is path traced), and its API and
output may still change; simple and advanced are the stable choices. See
Renderers.
Documentation: https://k3d-jupyter.org
Two frames from the renderers 3.0.0 added, both produced by the code in this repository — click either one for how it works.
advanced: image-based lighting and ambient occlusion. A million analytic sphere impostors,
with the occlusion in the crevices between strands doing the sculpting.
cinematic: progressive path tracing. A Draco-compressed glTF read with k3d.glb: the stones
refract with an index of 2.09 and throw light between their facets, the gold reflects the
studio around it. 256 samples, denoised with Open Image Denoise. Model: "Sasha" by saber7711,
CC-BY.
Click to watch at YouTube:
Watch: Interactive showcase gallery
To install from PyPI use pip:
$ pip install k3d
To install from conda-forge use:
$ conda install -c conda-forge k3d
Since 3.0.0 (the anywidget migration) no extra steps are needed:
!pip install k3d
import k3d and plot - custom widget activation and the text protocol
are no longer required.
To install directly from this repository (requires git and node.js + npm to build):
$ pip install git+https://github.com/K3D-tools/K3D-jupyter
This also makes possible installing the most up-to-date development version (same requirements):
$ pip install git+https://github.com/K3D-tools/K3D-jupyter@devel
To install any historical version, replace devel above with any tag or commit hash.
For a development installation (requires npm and node.js),
$ git clone https://github.com/K3D-tools/K3D-jupyter.git
$ cd K3D-jupyter
$ pip install -e .
No separate JupyterLab extension step is needed - the widget ships its own frontend module (anywidget).
K3D-jupyter follows the Python Software Foundation Code of Conduct in everything we do.
cinematic renderer traces its light with
three-gpu-pathtracer and
three-mesh-bvh by Garrett Johnson - thank you 🙏