constinit/imprint

The Imprint Project

Imprint

Imprint is a library to validate clinical trial designs.

Installing Imprint for development.

(Soon, we will have a separate pathway for users to install via PyPI/pip)

1. If you do not have conda installed already, please install it. There are many ways to get conda. We recommend installing Mambaforge which is a conda installation wwith mamba installed by default and set to use conda-forge as the default set of package repositories. CLICK HERE for installers and installation instructions.

2. Clone the git repo:

   git clone git@github.com:Confirm-Solutions/imprint.git

3. Set up your imprint conda environment. The list of packages that will be installed inside your conda environment can be seen in pyproject.toml.

   mamba update -y conda mamba
   # create a development virtual environment with useful tools
   mamba env create
   conda activate imprint
   # the following line makes poetry install into already activated
   # environment.
   poetry config virtualenvs.create false --local
   # install the imprint package plus development tools
   poetry install --with=dev,test
   
   # OPTIONALLY: Install a CUDA-enabled JAX build on Linux
   pip install --upgrade -r requirements-jax-cuda.txt

Getting started understanding imprint

• Tutorial: Z Test
• Tutorial: Chi-Squared Test
• Tutorial: T Test
• Tutorial: T Test Adaptive
• Tutorial: Fisher's Exact Test
• Tutorial: Three-arm Bayesian Basket Trial

References

• The main paper describing what imprint does: A Rigorous Framework for Automated Design Assessment and Type I Error Control: Methods and Demonstrative Examples
• Some older discussion of similar but obsolete methods can be found in Chapter 5 here: Adaptive Experiments and a Rigorous Framework for Type I Error Verification and Computational Experiment Design