Skip to content

Latest commit

 

History

History
26 lines (16 loc) · 590 Bytes

README.md

File metadata and controls

26 lines (16 loc) · 590 Bytes

Code for Generative Text Modeling through Short Run Inference

Pretrained Models

Pretrained models are available in cache/ckpt.

We also provide cached posterior z samples in cache. They can be used to evaluate the clustering and semi-supervised classification performance.

Run Trained Models

python train_sri.py

python unsupervised_cluster.py

python semi_sup_classification.py

Train New Models

python train_sri.py --eval_only False

Contact

Feel free to contact Bo Pang ([email protected]) or Erik Nijkamp ([email protected]) for questions.