Resource
BERT is a method of pre-training language representations which obtains state-of-the-art results on a wide array of NLP tasks.
Use the NGC CLI to download:
Copied!
Changelog
January 2023
- Pre-training using Language Datasets and Data Loaders (LDDL)
- Binned pretraining for phase2 with LDDL using a bin size of 64
August 2022
- Pre-training support with LAMB optimizer.
- Updated Data download and Preprocessing.
- Integrate DLLogger.
- Pre-training with AMP, Fleet.
- Pre-training support with Lamb optimizer.
- SQuAD finetune support with AdamW optimizer.
- Updated accuracy and performance tables tested on A100.
- Initial release.
Known issues
There are no known issues with this model.