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

Add support to load from any HF dataset and CSV #379

Merged
merged 3 commits into from
Feb 13, 2024
Merged

Add support to load from any HF dataset and CSV #379

merged 3 commits into from
Feb 13, 2024

Conversation

krypticmouse
Copy link
Collaborator

Code Snippet:

dl = DataLoader(
    train_size = 0.5,
    dev_size = 0.2,
    test_size = 0.3   
)

dl.from_huggingface("databricks/databricks-dolly-15k")

print(len(dl.train))
print(len(dl.dev))
print(len(dl.test))

arnavsinghvi11
arnavsinghvi11 reviewed Feb 13, 2024
View reviewed changes
dspy/datasets/dataloader.py
self._process_dataset(dataset, fields)

def from_csv(self, file_path:str, fields: List[str] = None):
dataset = load_dataset("csv", data_files=file_path)["train"]
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

just curious on why to keep ["train"] here. wondering if anyone has datasets not sorted to train/dev/test yet who may want to use it.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The idea was to support a single csv and create split from that. So the file_path would need to be a dict in order to support the multiple csv split. Though it's something we can iterate on!!

arnavsinghvi11
arnavsinghvi11 approved these changes Feb 13, 2024
View reviewed changes
Copy link
Collaborator

@arnavsinghvi11 arnavsinghvi11 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

looks good @krypticmouse ! really neat to have widely supported data loading!!

I actually wonder if we can deprecate hotpotqa and maybe refactor with this new DataLoader logic to keep it all clean. (but not needed, good as is too!)

@krypticmouse krypticmouse merged commit 7ea31fe into main Feb 13, 2024
1 check passed
@cognitivetech cognitivetech mentioned this pull request Feb 21, 2024
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@arnavsinghvi11 arnavsinghvi11 arnavsinghvi11 approved these changes

@okhat okhat Awaiting requested review from okhat

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@krypticmouse @arnavsinghvi11