namisan/mt-dnn

Multi-Task Deep Neural Networks for Natural Language Understanding

43
/ 100
Emerging

Built on BERT with shared encoder layers across multiple NLU tasks, the framework enables joint training on diverse datasets (GLUE, SQuAD, NER) while supporting adversarial training techniques like SMART and f-divergence regularization for improved robustness. Supports knowledge distillation, domain adaptation, sequence labeling, and question answering tasks with optimizations including gradient accumulation and FP16 training for efficient GPU utilization.

2,257 stars. No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 9 / 25
Community 24 / 25

How are scores calculated?

Stars

2,257

Forks

409

Language

Python

License

MIT

Last pushed

Mar 07, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/namisan/mt-dnn"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.