TEX2BEAM

This repository contains the tex2beam Python package that generates LaTeX Beamer presentations from LaTeX reports.

Requirements

Install the package requirements either using Python pip, or creating a Conda environment:

Python pip:

pip install -r requirements.txt

Conda:

conda env create -f environment.yml -n {ENVIRONMENT NAME}

Installation

The package can be installed by cloning this GitHub repository and installing the package locally:

pip install .

or directly from GitHub:

pip install git+https://github.com/TorStava/tex2beam.git

To install the package in development (editable) mode, use the following command in a local clone of the repository:

pip install --editable .

OpenAI API Token

All methods that access the OpenAI API requires a valid API TOKEN to be set in the .env file, or given as input parameter on the command line, or as input to the specific function. See the docstrings for the relevant functions for more details.

Getting Started

Loading the module for use in your own code:

import tex2beam as tb

Generating Presentations

To generate a presentation from the report $\LaTeX$ source file run the tex2beam.main module in the terminal, e.g.:

python -m tex2beam.main --latex-path {LATEX_PATH} --output-path {OUTPUT_PATH} --method "rag"

To generate presentations from all report source file in a folder run the main tex2beam script in the terminal, e.g.:

python -m tex2beam.main --source-folder {SOURCE_FOLDER} --target-folder {TARGET_FOLDER} --method "rag"

NOTE: Be careful if you run this on a large dataset as the conversions can consume a lot of tokens, which will incur costs.

Available command line options can be viewed in the help output:

python -m tex2beam.main --help

See the code docstrings for more information.

Metrics

To calculate the metrics for a generated presentation against a reference presentation, use the following command:

python -m tex2beam.run_metrics --scoring-method {bert, rouge} \\
    --predictions-file {PREDICTIONS_FILE} \\
    --reference-file {REFERENCE_FILE} \\
    --output {OUTPUT_FILE}

Available command line options can be viewed in the help output:

python -m tex2beam.run_metrics --help

See the code docstrings for more information.