Skip to content

Latest commit

 

History

History
72 lines (50 loc) · 1.74 KB

INSTALL.md

File metadata and controls

72 lines (50 loc) · 1.74 KB

Installation

The installation guide is inherited from StreamDiffusion.

The initial release only tested the Linux env with conda/Docker enviroment.

Step0: clone this repository

git clone https://github.com/Jeff-LiangF/streamv2v.git

Step1: Make Environment

You can install StreamV2V via conda, or Docker(explanation below).

# Using conda (Recommended)
conda create -n streamv2v python=3.10
conda activate streamv2v

Step2: Install PyTorch and other dependencies

Select the appropriate version for your system. Check the Pytorch doc.

# CUDA 11.8
pip3 install torch==2.1.0 torchvision==0.16.0 xformers --index-url https://download.pytorch.org/whl/cu118

# CUDA 12.1
pip3 install torch==2.1.0 torchvision==0.16.0 xformers --index-url https://download.pytorch.org/whl/cu121

Install other dependencies

pip install -r requirements.txt

Step3: Install StreamV2V

For Developer (If you want to modify codes)

python setup.py develop easy_install streamv2v[tensorrt]
python -m streamv2v.tools.install-tensorrt

For User

# for Latest Version (recommended)
pip install git+https://github.com/Jeff-LiangF/streamv2v.git@main#egg=streamv2v[tensorrt]
# Install TensorRT extension
python -m streamv2v.tools.install-tensorrt

(Only for Windows) You may need to install pywin32.

pip install --force-reinstall pywin32

Docker Installation (TensorRT Ready)

git clone https://github.com/Jeff-LiangF/streamv2v.git
cd streamv2v
docker build -t streamv2v:latest -f Dockerfile .
docker run --gpus all -it -v $(pwd):/home/ubuntu/streamv2v streamv2v:latest