Morpheme segmentation for Russian word forms via convolutional neural networks

Create and activate a virtual environment for python 3

python3 -m venv venv
. venv/bin/activate

Install requirements in your virtual environment

pip install -r requirements.txt

Train a model

python neural_morph_segm.py config/train_config.json

Test a trained model on a test set

python neural_morph_segm.py config/test_config.json

Repository structure

• сonfig: configuration files
• data: train and test sets (morpheme segmentations for word forms of Tikhonov's morphological dictionary).
• models: trained models
• neural_morph_segm.py: main file with model code
• read.py: input data reader