Company Classifier

Motivation

This is a project to classify companies based on their financial data.

Contributing

No contribution is needed at the moment.

Usage

Quick Start

This project only uses:

• python -> 3.11
• pandas -> 2.0.1
• python-dateutil -> 2.8.2
• numpy -> 1.26.4
• pyyaml -> 6.0.2
• pytest -> 8.3.5

Therefore, it's possible to use a local python environment and, run:

python main.py

To create the python environment you can run:

python -m venv ./.venv

And then the following, to activate it:

source ./.venv/bin/activate

Also, feel free to use any other environment manager, e.g.: pyenv or any other that you like.

You can also use docker compose, the option for compose is to facilitate the volume emulation inside the container, and any dynamic changes, not having to rebuild the container and copying the files all over again.

To run on Docker:

docker compose up

The above command will run the script once and save the file on your local directory, since the entire directory is being emulated inside the container.

Important

Don't forget to add your input/output directory, and to add your dataset!

How to configure this project

There are three important files:

• config.yaml
• rules.yaml
• .env

The first two are included with functional data in it, since they do not have any sensitive data. The .env file needs to be created locally, even it the parameters are empty.

Please take a brief look at their specific documentation:

• How to set up the config.yaml file
• How to set up the rules.yaml file