Installation

Install from GitHub

Clone the public repository and install the core package with a lightweight local backbone:

git clone https://github.com/yuelyu21/SurvFM.git
cd SurvFM
python -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install -e ".[sklearn]"

For a non-editable installation directly from GitHub:

python -m pip install "survfm-rmst[sklearn] @ git+https://github.com/yuelyu21/SurvFM.git"

Run the environment diagnostic without exposing credential values:

survfm-rmst diagnose
survfm-rmst backbones

Optional installations

Install only the environment needed for a backbone:

python -m pip install -e ".[tabpfn]"
python -m pip install -e ".[tabicl]"
python -m pip install -e ".[tabdpt]"
python -m pip install -e ".[tabh2o]"

MITRA is large enough that a separate environment is recommended:

python -m venv .venv-mitra
source .venv-mitra/bin/activate
python -m pip install -e ".[mitra]"

Model access

Some foundation-model checkpoints require license acceptance, login or a first-use download. Follow each model provider's official instructions and pin the package and checkpoint for a reproducible analysis.

Documentation environment

python -m venv .venv-docs
source .venv-docs/bin/activate
python -m pip install -e ".[docs]"
mkdocs serve --dev-addr 127.0.0.1:8000

The command serves a local documentation preview.