MOBO4EIC

a repo for developing MOBO tools for EIC.

The latest optimization pipeline can be monitored in weights and biases here

Installation

The following installation steps are followed

1. Install anaconda/miniconda from here
2. create an empty conda environment conda create --env ENV_NAME and activate conda activate ENV_NAME
3. After creating, install pip as conda install pip and run pip install pip_requirements.txt
4. Then continue with conda install --file conda_requirements.txt
5. To upload to the existing, project in weights & biases, make sure to sign up for an account, and let me know, so I can add teammates. Else, create a new project and modify temp_secrets.key to add the relevant API key

Usage

In order to run an optimization use the following command after loading the relevant environment in conda

>> cd MOBO-Closures
>> python wrapper.py -c optimize.config -p True -s secrets.key

This should run the optimization. Modify any hyperparameters for optimization in the optimize.config file.

To do

• Make the framework more modular. Seperate each of the data classes seperately
• Write more detailed documentation using git-book or jupyter-book