Saving and appending for loop results from every run

Question

LeoAiE 2021-8-30

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1443589-saving-and-appending-for-loop-results-from-every-run

评论： LeoAiE 2021-8-31

采纳的回答： Chunru

Test.xlsx

在 MATLAB Online 中打开

Hi everyone,

I have a dataset with countries and lat/lon columns. I want to measure the distance between the countries. I used the distance function from the mapping toolbox and a for loop to get every single country.

I’m struggling with saving the results from the for loop. I’m overwriting my results. I want to save every run from the for loop as a single column, basically append the results from every run. Does anyone have an efficient/better way to save the results? I would really appreciate your help in advance. I attached the data example and my code

format long g
Data = readtable('Test.xlsx');
for i=1:Data.lat
    Distance = deg2sm(distance([Data.lat(i) Data.lon(i)], [Data.lat(1:end) Data.lon(1:end)]));
    Data.Distance = Distance
end

2 个评论
显示无隐藏无

DGM 2021-8-30

编辑：DGM 2021-8-30

You appear to be measuring the distance from each point to all other points. The result won't fit in one column and won't be meaningfully aligned with the corresponding rows.

For example, let's assume you have 10 points. The way you're doing this, you'll have 10x10 = 100 distances. Of course, a lot of those distances will be redundant. There would only be 10!/2!(10-2)! = 45 unique distances. If your table is 10 rows long, where would you put 45 entries?

In the first case, you could add 10 distance columns to the table, each one listing the distance from one country to each country in the table. Whether you want to add a bunch of columns is up to you. Either way, it's not going to meaningfully fit in one column with the existing table length.

LeoAiE 2021-8-30

编辑：LeoAiE 2021-8-30

@DGM I agree with you it won't fit in one column. I was trying to save each run as a new column and I will deal witrh the rest later like puting it in a matrix where the column header and the row name is the country if that make sense. for now, any ideas how to save each run as a new column?

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Chunru 2021-8-31

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1443589-saving-and-appending-for-loop-results-from-every-run#answer_777429

在 MATLAB Online 中打开

Test.xlsx

If you want to attach the distance to the Data table you have read, then you can attach the distance vector for each country as follows.

format long g
Data = readtable('Test.xlsx');
for i=1:length(Data.lat)
    Distance = deg2sm(distance([Data.lat(i) Data.lon(i)], [Data.lat(1:end) Data.lon(1:end)]));
    Data.Distance{i} = Distance;
end
Data
Data = 19×4 table
latloncountryDistance__________________________________________________________

     42.546245       1.601554    {'Andorra'             }    {19×1 double}
     23.424076      53.847818    {'United Arab Emirates'}    {19×1 double}
      33.93911      67.709953    {'Afghanistan'         }    {19×1 double}
     17.060816     -61.796428    {'Antigua and Barbuda' }    {19×1 double}
     18.220554     -63.068615    {'Anguilla'            }    {19×1 double}
     41.153332      20.168331    {'Albania'             }    {19×1 double}
     40.069099      45.038189    {'Armenia'             }    {19×1 double}
     12.226079     -69.060087    {'Netherlands Antilles'}    {19×1 double}
    -11.202692      17.873887    {'Angola'              }    {19×1 double}
    -75.250973      -0.071389    {'Antarctica'          }    {19×1 double}
    -38.416097     -63.616672    {'Argentina'           }    {19×1 double}
    -14.270972    -170.132217    {'American Samoa'      }    {19×1 double}
     47.516231      14.550072    {'Austria'             }    {19×1 double}
    -25.274398     133.775136    {'Australia'           }    {19×1 double}
      12.52111     -69.968338    {'Aruba'               }    {19×1 double}
     40.143105      47.576927    {'Azerbaijan'          }    {19×1 double}

If you want to form a tall vector of distances, or a paired distance matrix you can do the following:

d = [];
for i=1:length(Data.lat)
    Distance = deg2sm(distance([Data.lat(i) Data.lon(i)], [Data.lat(1:end) Data.lon(1:end)]));
    d = [d; Distance];      % tall vector
    %d = [d  Distance];      % matrix
end
d
d = 361×1
                         0
          3243.52727137724
          3545.36144853792
          4082.36333070285
          4095.96109655197
          958.474221225425
          2236.15594470842
          4668.31071718444
          3854.00859782759
           8139.3428702947