Skip to content

RTP test file only has 100 prompts but the default rtp_sentences is 1000 #7

Description

@watate

Traceback (most recent call last):
File "/workspace/ml-act/act/scripts/pipeline.py", line 110, in main
EVAL_REGISTRYeval
File "/workspace/ml-act/act/scripts/pipeline.py", line 40, in _rtp
evaluate_toxicity.measure_toxicity(cfg.rtp)
File "/workspace/ml-act/act/evaluations/evaluate_toxicity.py", line 355, in measure_toxicity
) = evaluate_realtoxicityprompts_benchmark(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/workspace/ml-act/act/evaluations/evaluate_toxicity.py", line 159, in evaluate_realtoxicityprompts_benchmark
assert len(df_per_prompt) == n_prompts
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AssertionError

Sorry if it feels like I'm spamming issues! I just thought it would be good to report here as I go along. Let me know if you'd prefer for me to just report issues in my fork.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions