imgarylai/bert-embedding
A library that extracts token-level embeddings from pre-trained BERT models using MXNet and GluonNLP.
Not currently ranked — collecting fresh signals.
star history
This project extracts token-level embeddings from BERT’s pre-trained model weights. Instead of fine-tuning BERT end-to-end for downstream NLP tasks, users can extract and reuse the contextual token embeddings directly to build their own models. Implemented on Apache MXNet with the GluonNLP toolkit.
Frequently asked
- What is imgarylai/bert-embedding?
- A library that extracts token-level embeddings from pre-trained BERT models using MXNet and GluonNLP.
- Is bert-embedding open source?
- Yes — imgarylai/bert-embedding is open source, released under the Apache-2.0 license.
- What language is bert-embedding written in?
- imgarylai/bert-embedding is primarily written in Python.
- How popular is bert-embedding?
- imgarylai/bert-embedding has 451 stars on GitHub.
- Where can I find bert-embedding?
- imgarylai/bert-embedding is on GitHub at https://github.com/imgarylai/bert-embedding.