rl-based-news-rs

KO | EN

gitlite — search

#python #hacktoberfest #react #ai #go #machine-learning #rust #macos #nodejs #windows #awesome #kubernetes #pytorch #ios

rl-based-news-rs

(★ 14)

News Recommender System based on an Online Reinforcement Learning Model

rl-based-news-rs 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

<br>
<p align="center">
    <img src="images/logo-titled.png" width="40%" height="50%">
  <h3 align="center">News Recommender System with Online Reinforcement Learning</h3>
  </p>

## Table of Content
- [Table of Content](#table-of-content)
- [General Info](#general-info)
- [Features](#features)
- [Technologies](#technologies)
- [Setup](#setup)
- [Screenshots](#screenshots)
- [About Us](#about-us)
- [Acknowledgement](#acknowledgement)

## General Info
We implemented a News Recommender System with an Online Reinforcement Learning approach, based on [MIND](https://www.microsoft.com/en-us/research/publication/mind-a-large-scale-dataset-for-news-recommendation/) dataset.
Our first challenge was how to extract information from news texts and categories that represent precisely the data.
Second challenge was how to formulate problem in a RL literature. Finally, we had to figure out how to train the model in offline manner.

## Features
+ Designed a Deep Q-Network model using Pytorch
+ Deployed and tracked the model using the MLFlow tool
+ Developed a very straightforward back-end using FastAPI
+ Wrote a Dockerfile and dockerized the whole project

## Technologies
+ python = 3.9
+ pandas = 1.4.1
+ numpy = 1.22.2
+ nltk = 3.7
+ fastapi = 0.75.0
+ uvicorn = 0.17.5
+ mlflow = 1.24.0
+ torch = 1.11.0

## Setup
```console
# create dataset directories
md dataset
md dataset\train, dataset\dev, dataset\test

# download and extract train_set
wget -O train.zip 'https://mind201910small.blob.core.windows.net/release/MINDlarge_train.zip'
tar -xf train.zip -C dataset\train

# download and extract dev_set
wget -O dev.zip 'https://mind201910small.blob.core.windows.net/release/MINDlarge_dev.zip'
tar -xf dev.zip -C dataset\dev

# download and extract test_set
wget -O test.zip 'https://mind201910small.blob.core.windows.net/release/MINDlarge_test.zip'
tar -xf test.zip -C dataset\test

# remove dataset zips
rm train.zip, dev.zip, test.zip

# install dependencies
pip install poetry
rm poetry.lock
poetry install

# create features
python src/features.py
```

### Run Server with HTTP Protocol
```console
# run server
uvicorn run:app --proxy-headers --host '0.0.0.0' --port 8000

# run front-end
# example: user id is U687515
start .\front\code.html

# run mlflow
mlflow ui
```

### Run Server with HTTPS Protocol
```console
# run server
uvicorn run:app --proxy-headers --host '0.0.0.0' --port 8000 --ssl-keyfile './key.pem' --ssl-certfile './cert.pem'

# run front-end
# example: user id is U687515
start .\front\code.html

# run mlflow
mlflow ui
```

### Docker
Docker image is available at [this docker hub repository](https://hub.docker.com/repository/docker/haniehm26/recbrain)
```console
docker pull haniehm26/recbrain:latest-http
docker pull haniehm26/recbrain:latest-https

# port 8000 -> server
# port 5000 -> mlflow
docker run -p 5000:5000 -p 8000:8000 recbrain:latest-http
docker run -p 5000:5000 -p 8000:8000 recbrain:latest-https

# run front-end
# example: user id is U687515
start .\front\code.html
```

## Screenshots
<p align="center">
    <img src="images/front.gif" width="100%" height="100%">
  <h4 align="center">Project Front-End & Back-End Sample</h4>
  </p>
  <br>
<p align="center">
    <img src="images/mlflow.gif" width="100%" height="100%">
  <h4 align="center">Reward Cumulative Sum Metric Chart</h4>
  </p>
  <br>
  
## About Us
We are RecBrain, four data science enthusiasts:
+ [Hanieh Mahdavi](https://github.com/haniehm26)
+ [Mohammadamin Farajzadeh](https://github.com/mohamadamin-fjz)
+ [Mahta Zarean](https://github.com/mahtaz)
+ [Amir Asrzad](https://github.com/AMIRASRZAD)

Gather together in [Rahnema College](https://rahnemacollege.com/) Machine Learning Internship Program with mentorship of [Matin Zivdar](https://github.com/zivdar001matin).

## Acknowledgement
Special thanks to [Matin Zivdar](https://www.linkedin.com/in/matin-zivdar/) for his sincerely guidance and mentorship.<br>
Many thanks to [Rahnema College](https://rahnemacollege.com/) for their fantastic internship program, [Azin Azarkar](https://www.linkedin.com/in/azin-azarkar-8829b6183), [Yasin Orouskhani](https://www.linkedin.com/in/yasinorouskhani/), and everyone else who helped us through this project.

rl-based-news-rs

rl-based-news-rs 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

rl-based-news-rs

rl-based-news-rs 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README