### Install Uni-Core from Source

Source: https://github.com/dptech-corp/uni-core/blob/main/README.md

Instructions for building and installing Uni-Core directly from its source code. Users can choose to enable or disable CUDA extensions based on their environment's CUDA availability. Ensure the CUDA version in the build environment matches the one used by PyTorch.

```Shell
python setup.py install
```

```Shell
pip install .
```

```Shell
python setup.py install --disable-cuda-ext
```

--------------------------------

### Pull Uni-Core Docker Image

Source: https://github.com/dptech-corp/uni-core/blob/main/README.md

Command to pull the pre-built Uni-Core Docker image. To utilize GPUs within the Docker container, ensure nvidia-docker-2 is installed on the host system.

```Shell
docker pull dptechnology/unicore:0.0.1-pytorch1.11.0-cuda11.3
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.