Jump to content

Connect SuperML | Leeroopedia MCP: Equip your AI agents with best practices, code verification, and debugging knowledge. Powered by Leeroo — building Organizational Superintelligence. Contact us at founders@leeroo.com.

Principle:Facebookresearch Habitat lab Integration Testing

From Leeroopedia
Revision as of 17:37, 16 February 2026 by Admin (talk | contribs) (Auto-imported from principles/Facebookresearch_Habitat_lab_Integration_Testing.md)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Knowledge Sources
Domains Testing, Software_Architecture
Last Updated 2026-02-15 02:00 GMT

Overview

Validation of custom Habitat components by running them in a live environment to verify correct observation shapes, metric values, and action effects.

Description

Integration Testing verifies that custom sensors, measures, and actions work correctly when composed into a live Habitat environment. This involves creating an Env or HabGymWrapper with the custom components wired in, running episodes, and asserting that: sensor observations have the expected shapes and value ranges, measures produce valid metrics, and actions have the expected effects on simulation state.

Usage

Perform as the final validation step after defining, registering, and wiring custom components. Test with both habitat.Env (native API) and HabGymWrapper (Gym API) to ensure compatibility.

Theoretical Basis

Integration testing follows a standard pattern:

# Abstract integration test
env = create_env_with_custom_components(config)
obs = env.reset()
assert "my_sensor" in obs
assert obs["my_sensor"].shape == expected_shape

obs, reward, done, info = env.step(action)
assert "my_measure" in info["metrics"]
assert isinstance(info["metrics"]["my_measure"], float)

Related Pages

Implemented By

Page Connections

Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment