RepoPilot

physical-intelligence/openpi · Architecture

A static read of how physical-intelligence/openpi is wired together — module coupling, layering, and circular dependencies — derived from its import graph.

A
Architecture grade
score 88/100
2
Import cycles
lower is better
91
Modules parsed
130 edges
96%
Layeredness
clean-layer ratio

Most-depended-on modules

These are the highest-fan-in files — change them and the blast radius is largest. Read them first.

  • src/openpi/models/__init__.py12 importers · radius 24
  • src/openpi/shared/__init__.py11 importers · radius 24
  • src/openpi/models/model.py9 importers · radius 11
  • src/openpi/shared/array_typing.py8 importers · radius 23
  • src/openpi/training/config.py7 importers · radius 8
  • src/openpi/transforms.py7 importers · radius 11
  • src/openpi/models/pi0_config.py6 importers · radius 10
  • src/openpi/shared/download.py6 importers · radius 15

Modules (9)

Source directories grouped from the import graph. Each links to a per-module page with its files and how it couples to the rest.

Circular dependencies (2)

  • src/openpi/training/config.py to src/openpi/training/misc/polaris_config.py to src/openpi/training/config.py
  • src/openpi/training/config.py to src/openpi/training/misc/roboarena_config.py to src/openpi/training/config.py

Ask AI about physical-intelligence/openpi

Grounded in the actual source code. Pick a starter question or write your own.

Or write your own question

Informational only. RepoPilot summarises public signals at the time of analysis; they can be incomplete or stale. Not professional, security, or legal advice.