daptics_client - Python Client Package for the daptics API

daptics_client encapsulates the GraphQL daptics API, a powerful system for optimizing the design of experiments. Using daptics_client you can integrate the daptics iterative optimizations into lab automation processes.

daptics allows you to specify complex experimental design spaces, consisting of many numerical or categorical parameters, and uses highly-tuned machine learning algorithms to produce sequential batches of experimental designs (called "generations"), in an iterative process.

After performing the designed experiments in a batch and scoring the assay results of each experiment in the batch as a numeric response value, you upload these response values via the API to produce the next generation of designs.

As the process continues, daptics explores the experimental space to find potential optimal experiments for the next batch.

For a detailed look at the underlying mathematics, see this white paper: https://daptics.ai/pdf/White.pdf

Installation

Use setuptools to install the API modules on your Python system.

python3 setup.py install

Installation via pip coming soon!

Examples - Jupyter Notebooks

Example tutorial Jupyter notebooks and more information are available from the repository website at https://github.com/ProtoLife/daptics-api

Python Client and GraphQL API Reference Documentation

Documentation for both the client package and the low-level GraphQL API can be found at https://protolife.github.io/daptics-api

Web Access to the daptics System

A web interface to the system, together with more documentation and online help, is available at https://daptics.ai