القائمة
Encode game state as a Super Mini Map (SMM) with separate binary channels for players, ball, and ownership, bit-packed for efficient transfer in RL training
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Memory-efficient Keras generator that streams sharded CSV files in chunks, renders strokes to images on-the-fly, and yields batches for training on datasets too large for memory
Stack 1D convolutions for local feature extraction before bidirectional LSTMs to classify variable-length stroke sequences into hundreds of doodle categories
Partition a large dataset into N balanced shards using integer key modulo arithmetic for reproducible, class-interleaved splits across CSV files
Normalize raw stroke coordinates to 0-255 range, resample at uniform arc-length spacing, then apply Ramer-Douglas-Peucker simplification
Render stroke sequences to grayscale images with temporal intensity encoding where earlier strokes are brighter and later strokes fade to encode drawing order
Score each face in a video as anomalous by computing L2 distance from the embedding centroid of all faces, then convert to probability via logistic function
| name | cv-smm-spatial-observation-encoding |
| description | Encode game state as a Super Mini Map (SMM) with separate binary channels for players, ball, and ownership, bit-packed for efficient transfer in RL training |
For RL agents in spatial games (football, RTS), raw observation vectors (player positions as floats) lose spatial relationships. The Super Mini Map (SMM) renders game state onto a small 2D grid with separate binary channels: one for left team positions, one for right team, one for ball, one for active player. This gives CNN policies a spatial inductive bias while staying compact via bit-packing.
import numpy as np
from collections import deque
def render_smm(obs, width=96, height=72):
smm = np.zeros((height, width, 4), dtype=np.uint8)
def plot(positions, channel):
for x, y in positions:
px = int((x + 1) / 2 * (width - 1))
py = int((y + 0.42) / 0.84 * (height - 1))
px, py = np.clip(px, 0, width-1), np.clip(py, 0, height-1)
smm[py, px, channel] = 255
plot(obs["left_team"], 0)
plot(obs["right_team"], 1)
plot([obs["ball"][:2]], 2)
plot([obs["left_team"][obs["active"]]], 3)
return smm
# Frame stacking with bit-packing
frames = deque(maxlen=4)
obs = render_smm(raw_obs)
frames.extend([obs] * 4)
stacked = np.concatenate(list(frames), axis=-1)
packed = np.packbits(stacked, axis=-1)