Text -> Handwriting

Implementation

• Architecture: https://arxiv.org/pdf/2011.06704 repo
• Dataset: IAM On-Line Handwriting Database

Usage

1. Download the dataset from IAM, put it into the data directory
2. Make sure the python version is 3.12.0, use whatever environment manager (I prefer python -m venv env) and download requirements.txt
3. Run preprocessing.py, then run train.py