Skip to content

Updated readme. #9

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
joaomonteirof merged 1 commit into from
May 30, 2023
Merged

Updated readme. #9

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 13 additions & 6 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,13 +1,17 @@
# research-BertBigCode
Exploration of BERT-like models trained on The Stack
# BERT pre-training on The Stack
Exploration of BERT-like models trained on The Stack.

- Code used to train [StarEncoder](https://huggingface.co/bigcode/starencoder).
- StarEncoder was fine-tuned for PII detection to pre-process the data used to train [StarCoder](https://arxiv.org/abs/2305.06161)

**Work in progress. Currently training on the subsample of The Stack.**
- This repo also contains functionality to train encoders with contrastive objectives.

[Project information.](https://docs.google.com/document/d/1gjf7Y2Ek64xSyl8HE3GoK1kxDgsV8kjy-9pyIBkR-RQ/edit?usp=sharing)
- [More details.](https://docs.google.com/document/d/1gjf7Y2Ek64xSyl8HE3GoK1kxDgsV8kjy-9pyIBkR-RQ/edit?usp=sharing)


## To run locally:
## To launch pre-training:

After installing requirements, training can be launched via the example launcher script:

```
./launcher.sh
Expand All @@ -18,4 +22,7 @@ Exploration of BERT-like models trained on The Stack
- ```--train_data_name``` can be used to use to set the training dataset.

- Hyperparamaters can be changed in ```exp_configs.py```.
- The tokenizer to be used is treated as a hyperparameter and then must also be set in ```exp_configs.py```
- The tokenizer to be used is treated as a hyperparameter and then must also be set in ```exp_configs.py```.
- alpha is used to weigh the BERT losses (NSP+MLM) and the contrastive objective.
- Setting alpha to 1 corresponds to the standard BERT objective.
- Token masking probabilities are set as separate hyperparameters, one for MLM and another one for the contrastive loss.