我想获取直方图 bin 中包含的数据列表。我正在使用 numpy 和 Matplotlib。我知道如何遍历数据并检查 bin 边缘。但是,我想为 2D 直方图执行此操作,并且执行此操作的代码相当难看。numpy 是否有任何结构可以使这更容易?
对于一维情况,我可以使用 searchsorted()。但是逻辑并没有那么好,我真的不想在不需要时对每个数据点进行二进制搜索。
大多数讨厌的逻辑是由于 bin 边界区域造成的。所有区域都有这样的边界:[左边缘,右边缘)。除了最后一个 bin,它有一个像这样的区域:[left edge, right edge]。
以下是一维案例的一些示例代码:
import numpy as np
data = [0, 0.5, 1.5, 1.5, 1.5, 2.5, 2.5, 2.5, 3]
hist, edges = np.histogram(data, bins=3)
print 'data =', data
print 'histogram =', hist
print 'edges =', edges
getbin = 2 #0, 1, or 2
print '---'
print 'alg 1:'
#for i in range(len(data)):
for d in data:
if d >= edges[getbin]:
if (getbin == len(edges)-2) or d < edges[getbin+1]:
print 'found:', d
#end if
#end if
#end for
print '---'
print 'alg 2:'
for d in data:
val = np.searchsorted(edges, d, side='right')-1
if val == getbin or val == len(edges)-1:
print 'found:', d
#end if
#end for
以下是 2D 案例的一些示例代码:
import numpy as np
xdata = [0, 1.5, 1.5, 2.5, 2.5, 2.5, \
0.5, 0.5, 0.5, 0.5, 1.5, 1.5, 1.5, 1.5, 1.5, 2.5, 2.5, 2.5, 2.5, 2.5, 2.5, \
0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 1.5, 1.5, 1.5, 1.5, 1.5, 1.5, 1.5, 1.5, 2.5, 2.5, 2.5, 2.5, 2.5, 2.5, 2.5, 2.5, 3]
ydata = [0, 5,5, 5, 5, 5, \
15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, \
25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 30]
xbins = 3
ybins = 3
hist2d, xedges, yedges = np.histogram2d(xdata, ydata, bins=(xbins, ybins))
print 'data2d =', zip(xdata, ydata)
print 'hist2d ='
print hist2d
print 'xedges =', xedges
print 'yedges =', yedges
getbin2d = 5 #0 through 8
print 'find data in bin #', getbin2d
xedge_i = getbin2d % xbins
yedge_i = int(getbin2d / xbins) #IMPORTANT: this is xbins
for x, y in zip(xdata, ydata):
# x and y left edges
if x >= xedges[xedge_i] and y >= yedges[yedge_i]:
#x right edge
if xedge_i == xbins-1 or x < xedges[xedge_i + 1]:
#y right edge
if yedge_i == ybins-1 or y < yedges[yedge_i + 1]:
print 'found:', x, y
#end if
#end if
#end if
#end for
有没有更清洁/更有效的方法来做到这一点?似乎 numpy 会为此提供一些东西。