La interfaz está localizada; las descripciones y documentos de las Skills se traducen automáticamente con IA a su idioma. El código, los comandos, rutas y términos técnicos se mantienen en el original, y prevalece el documento del autor.
Explorar Skills
Datos e investigación · SKILL.md

evo

evo

Routes evo trajectory-evaluation workflows for APE/RPE, trajectory I/O, result analysis, configuration, and programmatic usage.

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentación de la Skill

GitHub

name: evo description: "Routes evo trajectory-evaluation workflows for APE/RPE, trajectory I/O, result analysis, configuration, and programmatic usage." disable-model-invocation: true metadata: disco-role: operating license: GPL 3.0

evo

Use this skill for the evo Python package, which evaluates odometry and SLAM trajectories, loads and converts trajectory files, compares saved results, manages package settings, and supports programmatic plotting and notebooks.

Install

  • Python: 3.10 or newer.
  • Install from a checkout with pip install -e ..
  • Install from PyPI with pip install evo.
  • Optional extras only when you need them:
    • pip install evo[rerun] for Rerun workflows.
    • pip install evo[gui] for the Qt plot backend.
    • pip install evo[geo] for map tiles via contextily.

Minimal checks

  • Import check: python -I -c "import evo; print(evo.__version__)"
  • Safe helper: python scripts/check_env.py
  • CLI help: evo --help

Route map

User needGo here
APE / RPE scoring, alignment, delta selection, result zipssub-skills/metrics/SKILL.md
Trajectory file formats, converters, sync, align, export, bag / MCAP loadingsub-skills/trajectory-data/SKILL.md
Saved result comparison, merge, table export, label cleanupsub-skills/result-analysis/SKILL.md
evo_config, evo pkg, evo cat_log, evo_ipythonsub-skills/configuration/SKILL.md
Custom Python scripts, notebooks, plotting, Rerun, contextilysub-skills/python-api/SKILL.md

When to read the bundled references

Working rules

  • Use the sub-skill that matches the user’s primary workflow instead of mixing APE/RPE, trajectory I/O, result comparison, and configuration into one answer.
  • Treat trajectory and metric helpers as in-place mutators unless the reference says otherwise; copy inputs when you need to preserve originals.
  • Keep references and scripts self-contained inside this skill tree. Do not rely on the original checkout being present.
  • If the package or APIs change, compare the checkout against references/repo-provenance.md and refresh the skill when needed.
Skills guardadas