/metre_detection

Automatic Poetic Metre Detection in Czech Verse

Primary LanguageJupyter NotebookMIT LicenseMIT

Automatic Poetic Metre Detection

A source code repository to the article: "Automatic Poetic Metre Detection" by Kristýna Klesnilová, Karel Klouda, Magda Friedjungová, and Petr Plecháč. (still preprint, TBA)

If you find this repository useful, please cite above mentioned article.

Contents

  • resources: Resources used inside the implementation
    • corpusCzechVerse: Corpus of Czech Verse Git repository
    • emnlp2017-bilstm-cnn-crf: BiLSTM-CRF Git repository
  • src: Implementation
  • requirements.txt: Python packages installation file
  • .env: Environment variables file

Environment for running all jupyter notebooks inside the implementation part

  1. Save absolute path of this directory to WORKING_DIR environment variable inside .env file (create .env if not exists): echo "WORKING_DIR=/absolute/path/to/this/dir" > .env
  2. Create Python3 virtual environment: python3 -m venv venv
  3. Activate virtual environment: source venv/bin/activate
  4. Install required packages: pip3 install -r requirements.txt

When using this code as Git repository cloned from GitHub

  1. Initialize Git submodules located inside resources: git submodule init & git submodule update

When using this repository as regular non-Git folder

  1. Create directory resources: mkdir resources
  2. Change directory to resources: cd resources
  3. Clone respected Git directories:
    • git clone git@github.com:magdafriedjungova/emnlp2017-bilstm-cnn-crf.git
    • git clone git@github.com:versotym/corpusCzechVerse.git