MCPcopy Create free account
hub / github.com/CausalLearning/MultiRobustness / __init__

Method __init__

pgd_attack.py:145–161  ·  view source on GitHub ↗

Attack parameter initialization. The attack performs k steps of size a, while always staying within epsilon from the initial point.

(self, model, attack_config, x_min, x_max, grad, reps=1)

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers

nothing calls this directly

Calls 1

nameFunction · 0.85

Tested by

no test coverage detected