Skip to content


Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time

daptics API

This is the README documentation for the daptics Design of Experiments GraphQL API and the Python API client.


To use the daptics API, you must first register at the to establish your login and password for API authentication.

Python Client

The python_client folder contains the Python GraphQL client package module sources (in the daptics_client, phoenix, and syncasync folders), and several interactive Python notebooks for experimenting with the API.

Follow the instructions in the file in that folder to set up a local Jupyter Notebook server if you do not have access to a server that can open .ipynb files.

GraphQL API Documentation

  1. Install graphql-markdown

  2. Then in the pydocmd folder, run:

NODE_TLS_REJECT_UNAUTHORIZED=0 graphql-markdown --no-toc --title 'Daptics GraphQL API' >


  1. The NODE_TLS_REJECT_UNAUTHORIZED=0 is to handle our ZeroSSL certificates.

  2. If this fails, you can use the appropriate JSON schema file to generate the docs, with this command:

graphql-markdown --no-toc --title 'Daptics GraphQL API' api-0.14.1.json >
  1. If using yarn to manage node packages, try this:
yarn bin graphql-markdown

to get the location of the graphql-markdown executable.

Python Client Documentation and MkDocs Build

  1. Install these tools in order:

    a. tornado - Important! Specify version 5.1.1 (version 6.0 will break MkDocs) b. pdoc3 c. MkDocs d. mkdocs-rtd-dropdown - Theme for MkDocs

  2. Create Markdown documentation for the file using pdoc3. In the python_client folder, run: pdoc --pdf --force --template-dir ../pdoc/templates daptics_client >../pydocmd/

  3. Then build the entire "Read the Docs" site, using mkdocs. In the root project folder, where the mkdocs.yml configuration file is located, run: mkdocs build

Html and Markdown files will be produced in the docs folder.

Using Jupytext to Extract and Sync to Python Source Files

  1. Install jupytext

  2. Set up metadata in any ipynb file that has Python code: jupytext --set-formats ipynb,python//py:light 03_SimpleTutorial.ipynb

  3. Export Python code to /python subdirectory: jupytext --from ipynb --to python//py:light 03_SimpleTutorial.ipynb

  4. Edit Python code as needed. Or when you run in Jupyter notebook, and make changes, the corresponding Python file will be kept up to date!

  5. Rebuild notebook from Python file without outputs (do this before checking into version control): jupytext --from python//py:light --to notebook python/

You can also use jupyter nbconvert to remove all output from .ipynb files:

jupyter nbconvert 03_SimpleTutorial.ipynb --to notebook --ClearOutputPreprocessor.enabled=True --inplace

Automated Release Notes by gren