https://github.com/carlobortolan/b1o.co
Chess-inspired algorithm for image ranking
https://github.com/carlobortolan/b1o.co
actix-web elo-rating mlp nextjs perl5
Last synced: about 1 month ago
JSON representation
Chess-inspired algorithm for image ranking
- Host: GitHub
- URL: https://github.com/carlobortolan/b1o.co
- Owner: carlobortolan
- License: gpl-3.0
- Created: 2024-05-01T20:29:09.000Z (over 1 year ago)
- Default Branch: master
- Last Pushed: 2025-03-13T22:16:06.000Z (7 months ago)
- Last Synced: 2025-04-11T19:55:44.242Z (6 months ago)
- Topics: actix-web, elo-rating, mlp, nextjs, perl5
- Language: Rust
- Homepage: https://2me-ai.vercel.app
- Size: 764 KB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# [b1o.co](https://b1o.co/about)
A Full-Stack Rust application where users can vote for their favorite images. Inspired from the [Elo rating system](https://en.wikipedia.org/wiki/Elo_rating_system#Theory) used in chess, each image is assigned a rating. A [Multi-Layer Feed-Forward Network](https://en.wikipedia.org/wiki/Feedforward_neural_network#Mathematical_foundations) is then used to analyze voting patterns and image characteristics to predict accurate base rankings for new images.
> [!WARNING]
> At the moment, the API is deployed on render ([**_https://api.b1o.co_**](https://api.b1o.co)) which can lead to significant loading times when opening the web-app.
## OVERVIEW
- [`/backend`](/backend): Rust API
- [`/client`](/client): Next.js client
- [`/crawler`](/crawler): Perl script to scrape data (following a similar concept to [PageRank's web-indexing](https://en.wikipedia.org/wiki/Web_crawler#Overview))
- [`/database`](/database): database models, migrations and seeds
- [`/cnn`](/cnn): classifier that uses the pre-trained [ResNet50](https://pytorch.org/hub/nvidia_deeplearningexamples_resnet50/) model to classify and post images
- [`/shared`](/shared): resources used by two or more components
## CONFIG
Start backend
`cargo build`
`cargo install cargo-make`
`cargo run --bin backend`
**OR**
`cargo make start_back_prod`
Start frontend
(`npm install`)
(`npm install -g pnpm`)
(`npm install -g wasm-pack`)
`pnpm install --dir ./frontend`
`pnpm run --dir ./frontend build`
`pnpm run --dir ./frontend dev`
**OR**
`cargo make start_front_prod`
**OR**
`simple-http-server ./frontend/dist -i -p 8000 --nocache --try-file ./frontend/dist/index.html`
> [!NOTE]
> You might have to `Set-ExecutionPolicy RemoteSigned` to run pnpm commands on Windows.
## RUNNING ON DOCKER
First install Docker and docker-compose.
You can find instructions [here](https://docs.docker.com/get-docker/) and [here](https://docs.docker.com/compose/install/).
In theory it would be sufficient to run `docker-compose up` and, once everything is up and running, visit http://localhost:3000 to view the app.
> [!NOTE]
> docker-compose will start the app in production mode.
To push to your remote docker-repository run:
```
docker-compose up
docker tag b1o-backend:latest /:backend
docker tag b1o-frontend:latest /:frontend
docker push /:backend
docker push /:frontend
```
## RUNNING OUTSIDE DOCKER
### Database Setup
Make sure to have a Postgres database with all read and write priviledges running.
Example:
```bash
sudo -u postgres psql
create database b1o;
create user coreuser with encrypted password 'password';
grant all privileges on database b1o to coreuser;
exit
```
> [!NOTE]
> If you get an error message saying "connection refused", your postgres server may have installed on port 5433
> rather than the default 5432. To resolve this, try using the -p flag as follows:
>
> ```bash
> sudo -u postgres psql -p 5433
> ```
### ENV file
You'll need a `.env` file to hold necessary environment variables.
If you're running docker, it should be named `docker.env` and contain the following:
```
FRONTEND_HOST=localhost
FRONTEND_PORT=3000
FRONTEND_PROTOCOL=http
DATABASE_PROTOCOL=postgres
DATABASE_PORT=5432
DATABASE_URL=
POSTGRES_USER=
POSTGRES_PASSWORD=
POSTGRES_DB=
POSTGRES_HOST=
BACKEND_HOST=localhost
BACKEND_HOST_INTERNAL=0.0.0.0
BACKEND_PORT=8080
BACKEND_PROTOCOL=http
MONGO_URL=
MONGO_COLLECTION=
```
### Running the App
[Cargo Make](https://github.com/sagiegurari/cargo-make) as a task running to simplify starting and stopping the
application.
To use cargo make, run `cargo install cargo-make`.
To start the application in development mode, run `cargo make start_all`.
To start the application in production mode, run `cargo make start_all_prod`.
> [!NOTE]
> On Windows, you might see the following error:
>
> ```powershell
> [webpack-dev-middleware] Error: spawn npm ENOENT
> at ChildProcess._handle.onexit (node:internal/child_process:285:19)
> at onErrorNT (node:internal/child_process:483:16)
> at process.processTicksAndRejections (node:internal/process/task_queues:82:21) {
> errno: -4058,
> code: 'ENOENT',
> syscall: 'spawn npm',
> path: 'npm',
> spawnargs: [ 'install', '-g', 'wasm-pack' ]
> }
> ```
>
> If you get this error, run the following command and then try again:
>
> ```
> npm install -g wasm-pack
> ```
To view all the available tasks, open [Makefile.toml](/Makefile.toml).
## CONTRIBUTING
If you find any bugs or have suggestions for improvement, please open a new issue or submit a pull request.
## LICENSE
This project is licensed under the GPL-3.0 license. See the [LICENSE](LICENSE) file for details.
> [!NOTE]
> The general concept has been inspired by [this](https://www.thecrimson.com/article/2003/11/4/hot-or-not-website-briefly-judges/) and [this](https://www.thecrimson.com/article/2003/11/19/facemash-creator-survives-ad-board-the/) article.
---
© Carlo Bortolan
> Carlo Bortolan ·
> GitHub [carlobortolan](https://github.com/carlobortolan) ·
> contact via [carlobortolan@gmail.com](mailto:carlobortolan@gmail.com)