Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/dmolitor/adaptive-infra
Infrastructure for adaptive learning with surveys and experiments.
https://github.com/dmolitor/adaptive-infra
Last synced: 7 days ago
JSON representation
Infrastructure for adaptive learning with surveys and experiments.
- Host: GitHub
- URL: https://github.com/dmolitor/adaptive-infra
- Owner: dmolitor
- License: gpl-3.0
- Created: 2023-12-04T22:25:39.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2024-07-16T17:57:33.000Z (5 months ago)
- Last Synced: 2024-11-01T22:50:34.231Z (about 2 months ago)
- Language: Python
- Size: 3.88 MB
- Stars: 0
- Watchers: 1
- Forks: 1
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# adaptive-infra
Want to host a survey but need it to be highly customizable, able to change its behavior in a data-adaptive manner,
and scalable (to a reasonable degree)? This repo shows how to build such a survey using
[Shiny for Python](https://shiny.posit.co/py/) and deploying/scaling it via AWS and
[Docker Swarm](https://docs.docker.com/engine/swarm/). This architecture is a part of a
[larger project](https://doi.org/10.17605/OSF.IO/CTD54)
demonstrating how researchers can use data-adaptive methods to detect discrimination in experimental settings more
efficiently than standard methods (randomized experiments and conjoint analyses).
## Requirements
- [AWS CLI v2](https://docs.aws.amazon.com/cli/latest/userguide/getting-started-install.html)
- [Configure the Cli for SSO](https://docs.aws.amazon.com/cli/latest/userguide/sso-configure-profile-token.html#sso-configure-profile-token-auto-sso)
- [just](https://github.com/casey/just)
- [Packer](https://developer.hashicorp.com/packer)
- [Python](https://www.python.org/downloads/)
### For local development
- [Docker Desktop](https://docs.docker.com/desktop/).
## Running locally
To get the stack running locally, it's as simple as:
```shell
docker compose up -d
```
To run using [Compose Watch](https://docs.docker.com/compose/file-watch/):
```shell
docker compose watch
```
To shut everything down:
```shell
docker compose down
```
## Services
The application is comprised of the following services:
- **app**:
This service builds and hosts the front-end survey form using [Shiny for Python](https://shiny.posit.co/py/). See
`/ui` for corresponding code.
- **api**:
This service builds and hosts the API that handles all interactions
between the survey form and the database on the backend using [FastAPI](https://fastapi.tiangolo.com/). See `/api`
for corresponding code.
- **database**:
This service utilizes the official PostgreSQL Docker image to store
all relevant data and user responses.
## Build/Deploy
This project uses [just](https://github.com/casey/just) to organize building/deployment
commands. To see all available recipes run `just` in the root directory.
```shell
>> just
just --list
Available recipes:
aws-ami-build # Build AWS EC2 AMI with HCL Packer
aws-security-group # Provision an AWS security group
aws-sso-configure # Configure AWS SSO
aws-sso-login profile="default" # AWS SSO login
aws-swarm-launch # Launch a Docker Swarm on an AWS EC2 server
aws-swarm-terminate # Terminate any active Docker Swarm AWS server
aws-volume # Provision an AWS EBS volume
black # Format with Black
check-aws # Check if AWS CLI is installed
check-dependencies # Check all deploy dependencies
check-docker # Check if Docker is installed
check-just # Check if just is installed
check-packer # Check if HCL Packer is installed
check-python # Check if Python is installed
default # List all available recipes
deploy # Deploy the application to an AWS-hosted server
docker-build-and-push # Build the app and api Docker images
prolific-studies # List all Prolific studies in the Adaptive Conjoint project
terminate # Terminate the running application and corresponding AWS server.
track-swarm-logs ip key service="api" # Track the Docker swarm logs
venv # Activate virtual environment and install Python dependencies
To execute a recipe: just [recipe-name]
```
### Build Docker images
To build the base Docker images and push them to DockerHub:
```shell
just docker-build-and-push
```
### Adaptive experiment
To deploy the adaptive experiment to an AWS cluster orchestrated by Docker Swarm:
```shell
just deploy
```
To terminate the running AWS cluster:
```shell
just terminate
```