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.

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

BoolQ for Training and Eval #30

Merged
merged 7 commits into from
Jun 21, 2024
Merged

BoolQ for Training and Eval #30

merged 7 commits into from
Jun 21, 2024

Conversation

farzadab
Copy link
Contributor

@farzadab farzadab commented Jun 14, 2024

This PR adds:

  • Training version of BoolQ (extended with GPT-based "explanations")
  • True/False (exact match) evaluations for BoolQ
  • Text-only evaluations for BoolQ and AnyInstruct to show the text-audio gap

ultravox/data/datasets.py Show resolved Hide resolved
ultravox/data/datasets.py Show resolved Hide resolved
ultravox/data/datasets.py Show resolved Hide resolved
ultravox/evaluation/eval_types.py Show resolved Hide resolved
ultravox/training/evaluation.py Show resolved Hide resolved
@farzadab farzadab marked this pull request as ready for review June 14, 2024 23:53
@farzadab farzadab requested a review from juberti June 14, 2024 23:53
@farzadab farzadab marked this pull request as draft June 17, 2024 16:10
@farzadab farzadab marked this pull request as ready for review June 19, 2024 23:44
ultravox/data/datasets.py Show resolved Hide resolved
ultravox/data/datasets.py Show resolved Hide resolved
ultravox/data/datasets.py Show resolved Hide resolved
ultravox/data/datasets.py Show resolved Hide resolved
ultravox/training/configs/stage2_lora.yaml Show resolved Hide resolved
ultravox/training/evaluation.py Outdated Show resolved Hide resolved
ultravox/training/evaluation.py Show resolved Hide resolved
ultravox/training/evaluation.py Outdated Show resolved Hide resolved
@farzadab farzadab merged commit 4202b56 into main Jun 21, 2024
1 check passed
@farzadab farzadab deleted the farzad-boolq-evals branch June 21, 2024 16:00
akshat0311 pushed a commit to jiviai/audio-llm that referenced this pull request Jan 30, 2025
* set default to include_context=True

* boolq extended dataset for training

* improved evals + boolq T/F eval + text-only
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants