This small project was carried out for the Deep Learning (Aprendizado Profundo) course integrating Unesp's Post-Graduate Program in Computer Science (Programa de Pós-Graduação em Ciência da Computação (PPGCC) da Unesp): https://www.ibilce.unesp.br/#!/pos-graduacao/programas-de-pos-graduacao/ciencia-da-computacao/apresentacao/
This course was taught by Professor Denis Henrique Pinheiro Salvadeo during the second semester of 2025.
The project was developed using Python 3.12.3 and the libraries (and versions) specified in requirements.txt in Ubuntu 24.04.3 / WSL
Clone this github repository and install the requirements with
pip install -r requirements.txt
Your google api key can be either provided to the respective summarizer class when initialized, to the pipeline when initialized or using the environmental variable API_KEY
export API_KEY="YOUR_API_KEY"
from src.pipeline import Beamifier_Pipeline
pipeline = Beamifier_Pipeline(model=model, api_key=api_key, _compile=True, remove_trash=True)
pipeline.run(input_path, output_path)
- model can either be the summarizer model's identifier (string) or its initialized class (check summarizers_dict inside summarizer.py for implemented models and identifiers)
- api_key can either be your api key (string) or set to None (if an api is used and api_key is set to None, it will attempt do gather it from the local variable API_KEY).
- _compile determines whether or not the pipeline attempts to compile the generated Beamer's .tex using pdfLaTeX
- remove_trash determines whether or not the pipeline deletes auxiliary files generated during the .tex file compilation if _compile is set to True
Distributed under the MIT License. See LICENSE for more information.
Project Link: https://github.com/filipipmp/Beamifier