メニュー
Score candidate movement directions by average distance to nearby opponents and pick the safest path for ball-carrying agents in game AI
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
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 | tabular-opponent-avoidance-scoring |
| description | Score candidate movement directions by average distance to nearby opponents and pick the safest path for ball-carrying agents in game AI |
When an agent controls the ball in a game AI, choosing which direction to advance requires evaluating opponent density ahead. Score each candidate direction (right, top-right, bottom-right) by the average distance to opponents in that sector. Pick the direction with the greatest average distance — or trigger a long pass if all directions are blocked (too many opponents nearby).
import numpy as np
def avg_distance_to_opponents(obs, target_x, target_y):
opponents = obs["right_team"]
dists = [np.sqrt((ox - target_x)**2 + (oy - target_y)**2) for ox, oy in opponents]
nearby = [d for d in dists if d < 0.05]
return np.mean(nearby) if nearby else 2.0, len(nearby)
def pick_direction(obs, player_x, player_y):
directions = {
"right": (player_x + 0.01, player_y),
"top_right": (player_x + 0.01, player_y - 0.01),
"bottom_right": (player_x + 0.01, player_y + 0.01),
}
best_dir, best_dist = "right", 0
total_nearby = 0
for name, (tx, ty) in directions.items():
dist, count = avg_distance_to_opponents(obs, tx, ty)
total_nearby += count
if dist > best_dist:
best_dist = dist
best_dir = name
if total_nearby >= 3:
return "high_pass" # surrounded — clear the ball
return best_dir