arxiv-scout

Semantic embedding-based search for the latest Arxiv papers

App Demo | Blogpost

Installation

1. Create a conda environment and pip install dependencies.

conda create -n arxiv-scout python=3.8
conda activate arxiv-scout
pip install -r requirements.txt

2. Sign up for a Cohere developer account and create a .env file at the project root directory. Initialize the environment variable COHERE_API_KEY with your Cohere API key. This is necessary for the dev environment. To run in production, set the environment variable COHERE_API_KEY to your Cohere API key via your production platform's secrets or environment variables management system.

# In .env file
COHERE_API_KEY = <YOUR_API_KEY_HERE>

3. Run notebooks in the /notebooks directory.

jupyter notebook

4. Run the Flask application to query

flask --app app:app run

Example request

POST /api/query

import requests
import json

url = "localhost:5000/api/query"

payload = json.dumps({
  "query": "Attention transformers are all you need"
})
headers = {
  'Content-Type': 'application/json'
}

response = requests.request("POST", url, headers=headers, data=payload)

print(response.text)

Project Organization

├── LICENSE
├── Makefile           <- Makefile with commands like `make data` or `make train`
├── README.md          <- The top-level README for developers using this project.
├── data
│   ├── external       <- Data from third party sources.
│   ├── interim        <- Intermediate data that has been transformed.
│   ├── processed      <- The final, canonical data sets for modeling.
│   └── raw            <- The original, immutable data dump.
│
├── docs               <- A default Sphinx project; see sphinx-doc.org for details
│
├── models             <- Trained and serialized models, model predictions, or model summaries
│
├── notebooks          <- Jupyter notebooks. Naming convention is a number (for ordering),
│                         the creator's initials, and a short `-` delimited description, e.g.
│                         `1.0-jqp-initial-data-exploration`.
│
├── references         <- Data dictionaries, manuals, and all other explanatory materials.
│
├── reports            <- Generated analysis as HTML, PDF, LaTeX, etc.
│   └── figures        <- Generated graphics and figures to be used in reporting
│
├── requirements.txt   <- The requirements file for reproducing the analysis environment, e.g.
│                         generated with `pip freeze > requirements.txt`
│
├── setup.py           <- makes project pip installable (pip install -e .) so src can be imported
├── src                <- Source code for use in this project.
│   ├── __init__.py    <- Makes src a Python module
│   ├── constants.py   <- Global constants that can be used by any module
│   ├── utils.py       <- Utility functions
│   │
│   ├── data           <- Scripts to download or generate data
│   │   └── arxiv_downloader.py <- Contains ArxivDownloader class to handle communication to Arxiv API
│   │
│   ├── features       <- Scripts to turn raw data into features for modeling
│   │
│   ├── models         <- Scripts to train models and then use trained models to make
│   │   │                 predictions
│   │   ├── cohere.py  <- CohereModel class to instantiate connections and make requests to Cohere API for embeddings
│   │ 
│   │
│   └── visualization  <- Scripts to create exploratory and results oriented visualizations
├── app.py             <- Flask microservice app that is deployed as the backend on Render
├── flask_config.py    <- Configurations for Flask app
└── tox.ini            <- tox file with settings for running tox; see tox.readthedocs.io

---

Citation

If used this work or you found this work useful, please use this BibTeX to cite this repository in your publications or works:

@software{jonathan_lim_siu_chi_2023_7595801,
  author       = {Jonathan Lim Siu Chi},
  title        = {jonathanlimsc/arxiv-scout: Arxiv Scout v1.0},
  month        = feb,
  year         = 2023,
  publisher    = {Zenodo},
  version      = {v1.0},
  doi          = {10.5281/zenodo.7595801},
  url          = {https://doi.org/10.5281/zenodo.7595801}
}

Project based on the cookiecutter data science project template. #cookiecutterdatascience