如何向量化Matlab代码？

Question

我想要渲染一些matlab代码。这是原始代码：

DelPos=find(q>I);

From=[0 DelPos]+1;
To=[DelPos I+J]-1;

L=cell(J,1);
UC=zeros(1,J);
for j=1:J
    L{j}=q(From(j):To(j));
    if ~isempty(L{j})                      
        UC(j)=sum(r(L{j}));
    end
end

这是我尝试对其进行矢量化：

DelPos=find(q>I);

From=[0 DelPos]+1;
To=[DelPos I+J]-1;

L=cell(J,1);
UC=zeros(1,J);

L{1:J}=q(From(1:J):To(1:J));
UC(1:J)=sum(r(L{1:J}));

但是，它会出错：

The right hand side of this assignment has too few values to satisfy the left hand side.

Error in ParseSolution (line 37)
    L{1:J}=q(From(1:J):To(1:J));

更新

我最终用2-dim数组替换了单元格，但是仍然无法对代码进行矢量化：

L=zeros(J,3000);
UC=zeros(1,J);
lenr=zeros(1,J);
for j=1:J
    ro=q(From(j):To(j));
    lenr(j)=length(ro);
    L(j,1:lenr(j))=ro;
    if ~isempty(L(j,1:length(ro)))                      
        UC(j)=sum(r(L(j,1:length(ro))));
    end
end

我的矢量化版本如下，但ro=q(From(1:J):To(1:J)）总是被最后一个J代替：

ro=q(From(1:J):To(1:J));
lenr(1:J)=length(ro(1:J));
L(1:J,1:lenr(1:J))=ro(1:J);                      
UC(1:J)=sum(r(L(1:J,1:length(ro(1:J)))));