Random IP

The random generator is a control baseline that ignores seed structure entirely. It simply emits cryptographically strong random bytes for each IPv6 address so you can quantify how much better a structured TGA performs on your measurement task.

• Reference: Placeholder citation (coming soon).

Train

rmap train --seeds seeds/active.txt --output models/random.bin random-ip

Training ignores the provided seeds and immediately produces a model wrapper that just records the configuration.

Generate

rmap generate --model models/random.bin --count 100000 \
  --output random-sample.txt

Configuration

• --seed <u64> – optional RNG seed. When omitted the generator seeds itself from the system RNG.

Model notes

• Use this baseline to compare hit rates against structured algorithms on the same scan budget.
• Because there is no learned structure, each generation run is independent, making it safe to run in parallel with different seeds.