This repo about analyze, generate and unify hints in current repo level coding problems.
folders - purpose:
analyze_hint: helper scripts of analyze hints, also a garbage collection of helper scriptshelper_code: helper code on the leave one out experiments, including data prepare and analysisidentity_projection: identity projection experiments given selection, currently empty, maybe merged intosrcidentity_selection: identity selection experiments given projection, currently has defuse2.py, feabench, nocode, and swebenchpro.leave_one_out: leave one out experiments, contains README inside.printing: instances printing scripts, contains READMErepos: location to store github repos for hint generation, contains scripts.src: extract identifier hints. contains outdated README, but suffice to see the functionsmini-swe-agent: changes to run mini swe agenttaxonomy_graph: the tools to draw the taxonomy graph.requirements.txtis meant for everything exceptmini-swe-agentandtaxonomy_graph.mini-swe-agentuses the environment from the officialmini-swe-agentgithub repo.taxonomy_graphfolder is self-contained.
Pull requests are very welcome! Make sure your patches are well tested. Ideally create a topic branch for every separate change you make. For example:
- Fork the repo
- Create your feature branch (
git checkout -b my-new-feature) - Commit your changes (
git commit -am 'Added some feature') - Push to the branch (
git push origin my-new-feature) - Create new Pull Request
All source files must include a Copyright and License header. The SPDX license header is preferred because it can be easily scanned.
If you would like to see the detailed LICENSE click here.
#
# Copyright IBM Corp. 2026
# SPDX-License-Identifier: MIT
#