fix: add memory-aware num_proc default in standardize_data_formats

[yurekami]

Summary

Previously, standardize_data_formats() defaulted to using all CPU cores (cpu_count()) when num_proc was not specified. This could exhaust memory when processing large datasets, as each worker imports heavy dependencies like PyArrow.

The fix adds memory-aware defaults:

• Starts with a reasonable base: min(max(psutil.cpu_count()+4, 2), 64)
• Checks available memory and limits workers accordingly (~1GB per worker)
• Falls back to 1 worker when memory is below 4GB
• Matches the safer approach already used in train_on_responses_only

Related Issue

Fixes #87

Test plan

• Test with large datasets (100k+ examples)
• Test on systems with low memory
• Test with explicit num_proc parameter (should override defaults)
• Verify memory usage stays reasonable during processing

🤖 Generated with Claude Code

[gemini-code-assist]

Summary of Changes

Hello @yurekami, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request introduces a crucial enhancement to the standardize_data_formats() function by implementing a memory-aware default for the num_proc parameter. Previously, the function risked exhausting system memory by defaulting to all available CPU cores, especially with large datasets. The updated logic dynamically adjusts the number of parallel workers based on real-time memory availability, significantly improving the robustness and stability of data processing operations, particularly on resource-constrained systems.

Highlights

• Memory-Aware num_proc Default: The standardize_data_formats() function now intelligently determines the default number of parallel processes (num_proc) by considering available system memory, in addition to CPU core count. This prevents Out-Of-Memory (OOM) errors that could occur when processing large datasets with many workers, each importing heavy dependencies like PyArrow.
• Dynamic Worker Allocation: The new logic calculates a base num_proc using psutil.cpu_count(), then limits it further based on available memory, assuming approximately 1GB per worker. If less than 4GB of memory is available, num_proc defaults to 1 to ensure stability.
• Consistency with Existing Practices: This change aligns the standardize_data_formats() function with the safer memory-aware approach already implemented in train_on_responses_only, promoting consistent resource management across the codebase.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request introduces a memory-aware default for num_proc in standardize_data_formats to prevent out-of-memory errors when processing large datasets. This is a significant improvement over the previous approach of defaulting to all available CPU cores. The new logic is more robust and considers available system memory to determine a safe number of worker processes. My review includes a few suggestions to further enhance code quality and maintainability, such as using isinstance for type checking, avoiding bare except clauses for better error handling, and ensuring consistency in calculations. Overall, this is a well-thought-out and valuable fix.

[gemini-code-assist]

Using a bare except: is a dangerous practice as it catches all exceptions, including system-exiting ones like SystemExit and KeyboardInterrupt, which can hide critical issues and make debugging difficult. It's better to catch a more specific exception, such as Exception.

Suggested change

	except:
	pass
	except Exception:
	pass

[gemini-code-assist]

It's more Pythonic to use isinstance() for type checking rather than comparing types directly with type(). This is more robust as it correctly handles subclasses.

Suggested change

	if num_proc is None or type(num_proc) is not int:
	if num_proc is None or not isinstance(num_proc, int):

[gemini-code-assist]

For consistency with other parts of the codebase (e.g., the train_on_responses_only function) and for improved readability, it's better to use (1024**3) for converting bytes to gigabytes.

Suggested change

	memory_gb_left = psutil.virtual_memory().available / 1024 / 1024 / 1024
	memory_gb_left = psutil.virtual_memory().available / (1024**3)