quale inspect

Loads a checkpoint file and prints the best genome’s topology. Optionally performs pathway analysis showing which sensors influence each actuator.

quale inspect is how you look inside an evolved brain to understand what emerged. After a run completes, it tells you which neurons exist, how they are connected, and - with --pathways - which sensors actually influence each actuator. This is where you find out whether the agent evolved the behavior you intended or found some other strategy entirely.

quale inspect <checkpoint-file> [flags]

Arguments

Argument	Description
<checkpoint-file>	Path to a .quale-ckpt checkpoint file

Flags

Flag	Default	Description
--pathways	false	Show sensor-to-actuator pathway analysis using reverse BFS

Output

Without --pathways:

=== Checkpoint: Generation 500 ===
Population size: 200

=== Best Genome ===
ID:         a1b2c3d4
Fitness:    42.50
Generation: 498
Nodes:       18 total (14 input, 2 hidden, 2 output)
Connections: 12 total (10 enabled, 2 disabled)
Sensors:     [signal_aspect distance_to_signal current_speed ...]
Actuators:   [throttle brake emergency_brake acknowledge_aws attention]

With --pathways:

--- Hidden Nodes ---
  Node 16: activation=Sigmoid bias=0.3421
  Node 17: activation=ReLU bias=-0.1234

--- Enabled Connections ---
  IN:signal_aspect(0) -> H(16)        weight=0.8234  inn=42
  H(16) -> OUT:throttle(14)           weight=1.2345  inn=56

--- Output Node Biases ---
  throttle: bias=0.1234
  brake: bias=-0.5678

--- Pathway Analysis ---
Sensors that influence each actuator (direct + indirect paths):
  throttle: current_speed, signal_aspect, speed_limit
  brake: braking_distance, signal_aspect
  emergency_brake: no connected sensors (output driven by bias only)

Examples

-- Basic topology inspection
quale inspect checkpoints/signalcompliance/checkpoint_gen500.quale-ckpt

-- Full pathway analysis
quale inspect checkpoints/signalcompliance/checkpoint_gen500.quale-ckpt --pathways

---

quale replay

Runs a single scenario with replay recording and outputs tick-by-tick brain activity as JSON to stdout. The replay captures sensor inputs, actuator outputs, and every node’s activation value at each step.

Requires a .quale project file (--quale) to configure the GenericDomain for the scenario.

quale replay <checkpoint-or-brain> --quale <file> [--seed <N>] [--ticks <N>]

Arguments

Argument	Description
<checkpoint-or-brain>	Path to a .quale-ckpt or .quale-brain file containing the genome to replay

Flags

Flag	Default	Description
--quale <file>	(required)	Path to the .quale project file that defines the simulation
--seed <N>	42	Random seed for the scenario
--ticks <N>	300	Maximum simulation ticks

Examples

-- Replay best brain from a checkpoint
quale replay checkpoints/signalcompliance/best.quale-brain \
    --quale rail-driver.quale --ticks 66000

-- Replay with a specific seed
quale replay checkpoints/signalcompliance/checkpoint_gen500.quale-ckpt \
    --quale rail-driver.quale --seed 123 --ticks 1000

---

quale export

Extracts or converts brain data from checkpoint or brain files. Supports three output modes: saving as a .quale-brain file, printing as JSON, or displaying a human-readable summary.

quale export <file> [--brain <path>] [--json]

Flags

Flag	Description
--brain <path>	Save/convert to a .quale-brain file at the given path
--json	Print brain topology as JSON to stdout

When no flags are given, prints a summary of the brain’s fitness, topology, and sensor/actuator names.

Examples

-- Print brain summary
quale export checkpoints/signalcompliance/best.quale-brain

-- Export checkpoint's best genome as a brain file
quale export checkpoints/signalcompliance/checkpoint_gen500.quale-ckpt --brain extracted.quale-brain

-- Export as JSON for external tooling
quale export checkpoints/signalcompliance/best.quale-brain --json