fix custom model example

[NathanHB]

No description provided.

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

[Copilot]

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.

[Copilot]

Pull Request Overview

This PR updates the custom model example to fix the configuration parameter name and refines the Google Translate example’s iteration and mapping.

• Rename model to model_name in CustomModelConfig instantiation.
• Update google_translate_model.py to use config.model_name and refactor dataset iteration.

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
src/lighteval/main_custom.py	Fixed parameter name passed to CustomModelConfig
examples/custom_models/google_translate_model.py	Mapped model_name correctly and switched to splits_iterator with nested loops

Comments suppressed due to low confidence (2)

examples/custom_models/google_translate_model.py:142

• [nitpick] The variable name r is ambiguous. Consider renaming it to a more descriptive name such as request or example for clarity.

for r in tqdm(split, desc="Batch", position=1, disable=False):

examples/custom_models/google_translate_model.py:139

• Hard-coding disable=False overrides the class-level disable_tqdm setting. Consider using disable=self.disable_tqdm to respect the configured flag.

disable=False,  # self.disable_tqdm,

[mjpost]

There are a few more instances:

• examples/custom_models/local_mt_model.py
• src/lighteval/models/endpoints/openai_model.py

[mjpost]

#768 for convenience