Cyrillic Mongolian text classification with tensorflow 2, and also some fine-tuning on TugsTugi's Mongolian BERT model and other NLP experiments are included.