本文介绍了在MATLAB中的textscan:读取NULL值为NaN的处理方法,对大家解决问题具有一定的参考价值,需要的朋友们下面随着小编来一起学习吧!
问题描述
我有一个.txt文件,包含以下数据:
$ b
sampleF.txt - >(制表符分隔)
I have a .txt file with the following data:
MSFT 200 100
APPL 10 NULL
AMZN 20 40
我需要使用 textscan
读取这些数据。我在阅读 NULL
数据时遇到问题。使用 treatasempty
param,我可以将其读为0.但是我想把它读作 NaN
。请帮忙!谢谢!
I need to read this data using textscan
. I am facing a problem while reading the NULL
data. Using treatasempty
param, I can read it as 0. But I want to read it as NaN
. Please help! Thanks!
fName = '.....\sampleF.txt'
[fid, message] = fopen(fName) ;
if fid < 0, disp(message), else
datatxt = textscan(fid, '%q %d %d', 'Delimiter', '\t','treatAsEmpty','NULL');
datatxt = [ datatxt {1} num2cell(datatxt {2}) num2cell(datatxt {3})] ;
fclose(fid) ;
end
%datatxt = { 'MSFT' [200] [100] ; 'AAPL' [10] [NaN] ; 'AMZN' [20] [40] }
推荐答案
问题在于类型 int32
不支持NaN值。而是把数字看作双打。即:
The problem is the fact that the type int32
does not support NaN values. Instead read the numbers as doubles. ie:
data = textscan(fid, '%s %f %f', 'Delimiter','\t', ...
'treatAsEmpty','NULL', 'EmptyValue',NaN);
这篇关于在MATLAB中的textscan:读取NULL值为NaN的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持!