giles 41adf058bd Build decord from source with CUDA for GPU video decode ...

- Build decord with -DUSE_CUDA=ON for true NVDEC hardware decode
- Use DLPack for zero-copy transfer from decord to CuPy
- Frames stay on GPU throughout: decode -> process -> encode

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

2026-02-04 01:50:14 +00:00

..

__init__.py

Replace batch DAG system with streaming architecture

2026-02-02 19:10:11 +00:00

arrays.py

Replace batch DAG system with streaming architecture

2026-02-02 19:10:11 +00:00

ascii.py

Replace batch DAG system with streaming architecture

2026-02-02 19:10:11 +00:00

blending_gpu.py

Add dev infrastructure improvements

2026-02-03 21:56:40 +00:00

blending.py

Replace batch DAG system with streaming architecture

2026-02-02 19:10:11 +00:00

color_ops_gpu.py

Add dev infrastructure improvements

2026-02-03 21:56:40 +00:00

color_ops.py

Add GPU frame conversion in color_ops

2026-02-03 21:38:10 +00:00

color.py

Replace batch DAG system with streaming architecture

2026-02-02 19:10:11 +00:00

core.py

Replace batch DAG system with streaming architecture

2026-02-02 19:10:11 +00:00

drawing.py

Replace batch DAG system with streaming architecture

2026-02-02 19:10:11 +00:00

filters.py

Replace batch DAG system with streaming architecture

2026-02-02 19:10:11 +00:00

geometry_gpu.py

Add dev infrastructure improvements

2026-02-03 21:56:40 +00:00

geometry.py

Replace batch DAG system with streaming architecture

2026-02-02 19:10:11 +00:00

image.py

Add IPFS HLS streaming and GPU optimizations

2026-02-03 20:23:16 +00:00

math.py

Replace batch DAG system with streaming architecture

2026-02-02 19:10:11 +00:00

streaming_gpu.py

Build decord from source with CUDA for GPU video decode

2026-02-04 01:50:14 +00:00

streaming.py

Disable GPU persistence by default

2026-02-03 21:24:45 +00:00