matplotlib

차이

문서의 선택한 두 판 사이의 차이를 보여줍니다.

차이 보기로 링크

--- matplotlib [2020/07/21 16:03] – rex8312
+++ matplotlib [2024/03/23 02:38] (현재) – 바깥 편집 127.0.0.1
@@ 줄 171: / 줄 171: @@
 </code>
-===== MAP-Elite archive =====
-<code python>
-behavior_description = np.random.randint(0, 3, (20, 4))
-quality = np.random.random(20)
-map_size = behavior_description.max(0) - behavior_description.min(0) + 1
-height_size = 1
-width_size = 1
-height_idx = []
-width_idx = []
-for idx, s in enumerate(map_size):
-	if height_size > width_size:
-    	width_idx.append(idx)
-    	width_size *= s
-    else:
-    	height_idx.append(idx)
-		height_size *= s
-height_ticks = map_size[height_idx]
-height = np.product(height_ticks)
-width_ticks = map_size[width_idx]
-width = np.product(width_ticks)
-bd_map = np.zeros((height, width))
- bd_idxs = behavior_description - behavior_description.min(0)
-for bd_idx, qv in zip(bd_idxs, quality):
-	y = 0
-	for i, hi in enumerate(bd_idx[height_idx]):
-    	y += hi * np.product(height_ticks[i+1:])
-	x = 0
-    for i, wi in enumerate(bd_idx[width_idx]):
-    	x += wi * np.product(width_ticks[i+1:])
-	bd_map[y][x] = qv
-fig = plt.figure()
-ax = fig.add_subplot(111)
-ax.set_xticklabels([])
-ax.set_yticklabels([])
-n_solutions = len(quality)
-search_space_size = np.product(bd_map.shape)
-ax.set_xlabel(f'{n_solutions:,} / {search_space_size:,} = {n_solutions/search_space_size:.3f}')
-im = ax.imshow(bd_map, cmap='jet', vmin=0, vmax=1)
-fig.colorbar(im)
-plt.show()
-plt.close()
-</code>
 ===== Neural Net Weights/Activation =====

matplotlib.1595347419.txt.gz · 마지막으로 수정됨: 2024/03/23 02:37 (바깥 편집)