feat: add custom output directory support to vf-eval#665
feat: add custom output directory support to vf-eval#665ericlau98 wants to merge 2 commits intoPrimeIntellect-ai:mainfrom
Conversation
Allow users to specify a custom output path via the -s/--save-results flag. When a path is provided, results are saved there instead of the default ./outputs or environment-local outputs directory. Usage: vf-eval gsm8k -s # default location vf-eval gsm8k -s /custom/path # custom location
There was a problem hiding this comment.
Config file references wrong environment ID
The math-python.toml configuration file sets id = "primeintellect/wiki-search" and run_name = "wiki-search", but it should reference the math-python environment instead. The environments/math_python/ directory exists with proper metadata indicating the environment ID is math-python. This appears to be a copy-paste error from another config file, causing the math-python config to load an entirely different environment when used.
configs/vf-rl/math-python.toml#L3-L4
verifiers/configs/vf-rl/math-python.toml
Lines 3 to 4 in 0e0e872
configs/vf-rl/math-python.toml#L17-L18
verifiers/configs/vf-rl/math-python.toml
Lines 17 to 18 in 0e0e872
The config file had copy-paste errors referencing wiki-search instead of math-python for both the environment ID and run_name.
00d66a2 to
cd4c172
Compare
|
Updated the math-python.toml based purely on what the Bugbot said, otherwise didn't touch that file. All existing functionality is still built in, nothing changes about default behaviour. I added the ability to define a custom output path using the -s/--save-results flag to allow for streamlining custom workflows. @willccbb Please let me know if you need anything else from me, would love to have this merged. |
|
Can ignore the bugbot comment, not related to the PR. What's the main use case for you for custom save paths? I get that it's maybe nice to have sometimes, but this might conflict with some other planned changes + not sure we're gonna want to maintain the plumbing here as-is. Definitely a nice feature idea though, wanna maybe open as an issue? Can make sure we incorporate something like this into upcoming updates. |
@willccbb The use case is for automation! The change maintains existing functionality should no custom path be specified. Are there any specific concerns I'm not thinking of? |
|
we're planning a more comprehensive update to vf-eval / related features to support better configurability + more robustness for building on top of it; currently it's intended as a pretty lightweight manual helper script will leave this open + think about it a bit more for a stopgap, would prefer if we can do this without the dual-use of -s and just have a more explicit flag |
Summary
-s/--save-resultsflag invf-eval./outputsoutput_dirfield toEvalConfigand updatespath_utils.pyUsage
Note
Enables saving eval outputs to a custom directory and adds a new RL config.
--save-results/-snow optionally accepts aPATH; without a path it behaves as a boolean flag. The chosen path is propagated asoutput_dir.output_dirthrough the stack: addoutput_dirtoEvalConfig, pass it fromverifiers/scripts/eval.py, and makeget_eval_results_pathinverifiers/utils/path_utils.pyhonor it while preserving default./outputsbehavior.configs/vf-rl/math-python.tomlwithenv.idandrun_nameset tomath-python.Written by Cursor Bugbot for commit cd4c172. This will update automatically on new commits. Configure here.