UniqAI/gaussian-splatting
1
0
Fork 0
mirror of https://github.com/graphdeco-inria/gaussian-splatting synced 2025-06-26 18:18:11 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
Sergio Bromberg 2023-09-05 12:34:37 +09:00 committed by GitHub
parent 0292f3c13d
commit 6423a7db39
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 12 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
README.md
View File

@ -44,11 +44,21 @@ User [camenduru](https://github.com/camenduru) was kind enough to provide a Cola
## Docker
Use the Dockerfile to build a working docker image with this repository. In order to build the container succesfully make sure that:
Use the Dockerfile to build an (Ubuntu 22.04) image with this repository. In order to build the image succesfully make sure that:
- NVIDIA's drivers are updated in the host machine (tested with 525.125.06)
- The default-runtime for docker is set to "nvidia" in the /etc/docker/daemon.json file. See [this](https://github.com/NVIDIA/nvidia-docker/issues/1033).
- The default-runtime for docker is set to `nvidia` in the `/etc/docker/daemon.json` file. See [this](https://github.com/NVIDIA/nvidia-docker/issues/1033).
- For the above to work, the [nvidia-container-toolkit](https://gitlab.com/nvidia/container-toolkit/container-toolkit) should be installed in your host (tested with version 1.14.0-rc.2).
Build your docker image from the Dockerfile by downloading the [Dockerfile](https://github.com/sergiobd/gaussian-splatting-docker/blob/docker/Dockerfile) and placing it in an empty folder. You can do this manually or by using `wget`. Then, build the docker image and run it with GPU enabled.
```
mkdir gaussian_splatting
cd gaussian_splatting
wget https://raw.githubusercontent.com/sergiobd/gaussian-splatting-docker/docker/Dockerfile
build -t gaussian_splatting .
docker run -it --gpus all --name gaussian_splatting --shm-size=64g -v /your/local/shared/folder:/workspace/data gaussian_splatting:latest bash
```
Please refer to [Docker's documentation](https://docs.docker.com/) for specific docker instructions.
## Cloning the Repository
The repository contains submodules, thus please check it out with