DeScribble

Use Machine Learning (ML) on handwritten text to draw inferences about the gender of the person who wrote it.

Models used

The project uses a ViT Image Classifier Transformer model for predictions. It performs better than a standard Random Forest Image Classifier (baseline model).

Requirements and Environment

DVC

This project uses DVC. It is included as a requirement in the requirements.txt file.

Python Environment

For installing the virtual environment you can install it manually with the following commands:

For Linux / Mac OS:

pyenv local 3.11.3
python -m venv .venv
source .venv/bin/activate
pip install --upgrade pip
pip install -r requirements.txt

For Windows, please use the following commands instead:

pyenv local 3.11.3
python -m venv .venv
.venv\Scripts\Activate.ps1
pip install --upgrade pip
pip install -r requirements.txt

Usage

Overview

Put samples in images/bw
Run the script segmenter.py to cut out words.
Create a table index_raw.csv with the columns Geschlecht and Datei Index
Train the ViT Model with the script vit_train.py

Using Data Versioning with DCC

To download the relevant data using DVC, run the following command (Linux / Mac OS / Windows):

dvc pull

publicimageltd / DeScribble