MCPcopy Create free account
hub / github.com/rlcode/reinforcement-learning / _ortho

Function _ortho

3-atari/2-ppo.py:34–37  ·  view source on GitHub ↗
(layer, gain)

Source from the content-addressed store, hash-verified

32
33
34def _ortho(layer, gain):
35 nn.init.orthogonal_(layer.weight, gain)
36 nn.init.zeros_(layer.bias)
37 return layer
38
39
40# Nature CNN shared trunk + policy and value heads.

Callers 1

__init__Method · 0.70

Calls

no outgoing calls

Tested by

no test coverage detected