Skip to content

Latest commit

 

History

History
43 lines (29 loc) · 1.39 KB

installation.md

File metadata and controls

43 lines (29 loc) · 1.39 KB
Raw

Installation Guide

YOLO-World is built based on pytorch=1.11.0 and mmcv=2.0.0.

We provide the requirements files in ./requirements:

  • basic_requirements: training, finetuning, evaluation.
  • demo_requirements: running YOLO-World demos.
  • onnx_requirements: converting YOLO-World to ONNX or TFLite models (TFLite is coming soon).

Install MMCV

YOLO-World adopts mmcv>=2.0.0. There are several ways to install mmcv

1. using openmim:

see more in official guide.

pip install openmim
mim install mmcv==2.0.0

2. using pip:

go to install-with-pip to select the pip index.

# cuda=11.3, torch=1.11
pip install mmcv==2.0.0 -f https://download.openmmlab.com/mmcv/dist/cu113/torch1.11/index.html
# cuda=11.7, torch=1.13
pip install mmcv==2.2.0 -f https://download.openmmlab.com/mmcv/dist/cu117/torch1.13/index.html
# cuda=12.1, torch=2.1
pip install mmcv==2.1.0 -f https://download.openmmlab.com/mmcv/dist/cu121/torch2.1/index.html

3. using whl

go to index packages to find a suitable version and download.

pip install mmcv-2.0.1-cp38-cp38-manylinux1_x86_64.whl