EDSL: AI-Powered Research
Expected Parrot Domain-Specific Language (EDSL) is an open-source Python package for conducting AI-powered research.
EDSL is developed by Expected Parrot and available under the MIT License.
This page provides documentation, tutorials and demo notebooks for the EDSL package and the Coop: a platform for creating, storing and sharing AI research. The contents are organized into key sections to help you get started.
Researchers
Are you using EDSL for a research project? We’d love to hear about your experience!
Send us an email at info@expectedparrot.com and we’ll provide credits to run your project or a gift card for your time.
Links
Download the current version of EDSL at PyPI.
Get the latest EDSL updates at GitHub.
Create a Coop account to store and share your research and access special features, including:
Survey Builder: An interface for launching hybrid human-AI surveys
Remote Inference: Run surveys on the Expected Parrot server
Remote Caching: Automatically store results and API calls on the Expected Parrot server
File Store: Store and share data files for use in EDSL projects
Explore research at the Coop.
Join the Discord channel.
Introduction
Overview: An overview of the purpose, concepts and goals of the EDSL package.
Whitepaper: A whitepaper about the EDSL package (in progress).
Citation: How to cite the package in your work.
Papers: Research papers and articles that use or cite EDSL.
Technical Setup
Installation: Instructions for installing the EDSL package.
Coop: Create, store and share research on the Expected Parrot server.
API Keys: Instructions for storing API keys to use EDSL locally (optional).
Getting Started
Starter Tutorial: A tutorial to help you get started using EDSL.
Core Concepts
Questions: Learn about different question types and applications.
Scenarios: Explore how questions can be dynamically parameterized for tasks like data labeling.
Surveys: Construct surveys with rules and conditions.
Agents: Design AI agents with relevant traits to respond to surveys.
Language Models: Select language models to generate results.
Working with Results
Results: Access built-in methods for analyzing survey results as datasets.
Caching LLM Calls: Learn about caching and sharing results.
Exceptions & Debugging: Identify and handle exceptions in running surveys.
Token usage: Monitor token limits and usage for language models.
Coop
Coop is a platform for creating, storing and sharing EDSL content and AI-based research.
Coop: Learn how to create, store and share content at the Coop.
Survey Builder: An interface for launching hybrid human-AI surveys.
File Store: Store and share files for use in EDSL projects.
Remote Inference: Run surveys on the Expected Parrot server.
Remote Caching: Automatically store survey results and API calls on the Expected Parrot server.
Notebooks: Instructions for posting .ipynb files to the Coop.
Importing Data
Conjure: Automatically import other survey data into EDSL to:
Clean and analyze your data
Create AI agents for respondents and conduct follow-on interviews
Extend your results with new questions and surveys
Store and share your data on the Coop
How-to Guides
Examples of special methods and use cases for EDSL, including:
Data labeling
Data cleaning
Analyzing survey results
Adding data to surveys from CSVs, PDFs, images and other sources
Conducting agent conversations
Converting surveys into EDSL
Cognitive testing
Notebooks
Templates and example code for using EDSL to conduct different kinds of research. We’re happy to create a new notebook for your use case!
Developers
Information about additional functionality for developers.