MCPcopy
hub / github.com/tensorlayer/TensorLayer / prepro

Function prepro

examples/reinforcement_learning/tutorial_atari_pong.py:47–54  ·  view source on GitHub ↗

Prepro 210x160x3 uint8 frame into 6400 (80x80) 1D float vector.

(I)

Source from the content-addressed store, hash-verified

45
46
47def prepro(I):
48 """Prepro 210x160x3 uint8 frame into 6400 (80x80) 1D float vector."""
49 I = I[35:195]
50 I = I[::2, ::2, 0]
51 I[I == 144] = 0
52 I[I == 109] = 0
53 I[I != 0] = 1
54 return I.astype(np.float32).ravel()
55
56
57env = gym.make("Pong-v0")

Callers 1

Calls

no outgoing calls

Tested by

no test coverage detected

Used in the wild real call sites across dependent graphs

searching dependent graphs…