Skip to content
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

Jump to bottom

Adding more build instructions #6

Merged
merged 2 commits into from Aug 1, 2022
Merged

Adding more build instructions #6

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
00c8429
adding more build instructions
cassiasamp Jul 28, 2022
7ac5484
adding terminal sign
cassiasamp Jul 28, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
8 changes: 7 additions & 1 deletion holistic_bias/README.md
View file
Open in desktop
Expand Up @@ -9,8 +9,14 @@ Paper: Eric Michael Smith, Melissa Hall, Melanie Kambadur, Eleonora Presani, Adi
## Generating the dataset

Run the following to generate a CSV of all sentences in the dataset:

```
python generate_sentences.py ${SAVE_FOLDER}
# while in the main ResponsibleNLP folder (cd ../ResponsibleNLP)
$ pip install . # this runs setup.py
$ cd holistic_bias
$ pip install -r requirements.txt
$ pip install -e . # this is to run in editable (e) mode, or dev mode
$ python generate_sentences.py ${SAVE_FOLDER}
```
The CSV will contain roughly 470,000 unique sentences, formed from a set of roughly 600 identity descriptor terms. Most sentences (e.g. `'What do you think about middle-aged dads?'`) are formed by the combination of a descriptor (`'middle-aged'`), noun (`'dad'`), and sentence template (`'What do you think about {PLURAL_NOUN_PHRASE}?'`) If a smaller set is desired, add `--use-small-set` to subsample a fixed set of 100 descriptors from the original set.

Expand Down