Question

如何对每个细胞子部分的各列进行求和？给定一个单元格A

A = {'a'  '546.8'  '543.5'  '544'  
  'a'  '641.9'  '637.4'  '632.3'
  'a'  '214.7'  '214.1'  '231.8'
  'a'  '256.9'  '255.6'  '254.2'
  'c'  '356'    '355.1'  '354.4'
  'c'  '759'    '759.6'  '756.2'
  'c'  '352.2'  '350.4'  '350.8'
  'f'  '234'    '230.3'  '232.3'
  'f'  '225'    '223.5'  '221.8'}

根据A的第一列中的不同字母将A分成子单元格。并且对子单元格的每列进行求和。

预期结果是：

B = {'a'  '1660.3'  '1650.6'  '1662.3'  
  'c'  '1467.2'    '1465.1'  '1461.4'
  'f'  '459'    '453.8'  '454.1'}

Answer 1

不需要循环：

%// get unique rows
[ids,~,subs] = unique(A(:,1))
%// transform string data to numeric data
vals = str2double(A(:,2:end))
%// sum unique rows
sums = accumarray(subs, 1:numel(subs), [], @(x) {sum(vals(x,:),1)} )
%// output result
out = [ids(:),num2cell(cell2mat(sums))]

Answer 2

可能的解决方案之一是

[B,~,idxs]= unique(A(:,1))
for k=2:size(A,2)
    B(:,k)= num2cell(accumarray(idxs,str2double(A(:,k))))
end

根据包含数值数据作为字符串的行索引求和子矩阵

2 个答案: