cannot load csv file

Hi , I am new to matlab and I face this error each time I try to load my csv file using the dbload command in iris TOOLBOX.

d=dbload('data.csv')

d =

struct with no fields.

how can I solve this problem?

Answer 1

[Image Analyst]

Instead try using the functions: readtable, readmatrix, readcell, or csvread

If you have any more questions, then attach your data and code to read it in with the paperclip icon after you read this:

TUTORIAL: How to ask a question (on Answers) and get a fast answer

[Tahina Riantsoa]

And what if I have to use the dbload function ? And each time I upload my data it always shows ''struct no field''

[Walter Roberson]

If you must use dbload() but dbload() consistently returns no fields for your input files, then all you can do is detect the "no fields" case and do something to handle it smoothly (such as giving a warning message and going on to the next file.)

d = dbload('data.csv');
if ~isstruct(d) || isempty(fieldnames(d))
    %handle the case where d has no fields
else
    %do real work
end

dbload() does whatever dbload() does. There is always the possibility of editting dbload() to do something different than it currently does; it is not clear that would be acceptable.

If you were to attach a sample data file, we could examine it to try to determine why dbload() is saying "no fields"

[Tahina Riantsoa]

ok this is my file

[Image Analyst]

Try this:

d = importdata('data.csv')

d = struct with fields:
        data: [52x7 double]
    textdata: {53x9 cell}

numericalData = d.data;
% Get 7 column names.  Adjust column indexes as needed.
columnNames = d.textdata(1, 2:end-1)

columnNames = 1x7 cell array
    {'GDP'}    {'CPI_U'}    {'S'}    {'RS'}    {'GDP_RW'}    {'CPI_RW'}    {'RS_RW'}

% Convert to a table if you want to associate the variables with the columns.
t = array2table(numericalData, VariableNames = columnNames)

t = 52x7 table
    GDP     CPI_U      S       RS       GDP_RW      CPI_RW    RS_RW
    ____    _____    ______    ___    __________    ______    _____

    3712    51.39    2599.2     12    2.3562e+06    87.28      NaN 
    3643    51.44    2491.9     12    2.4117e+06    88.48      NaN 
    3712    51.46    2603.4     12    2.3907e+06    88.54      NaN 
    4908    53.48    2630.2     12      2.48e+06    89.65      NaN 
    4150    55.46    2631.1     12    2.4021e+06    90.21      NaN 
    3727    55.55    2498.7     12     2.455e+06    91.69      NaN 
    4016    57.06    2366.7     12    2.4095e+06    91.94      NaN 
    5153    59.02    2589.1     12    2.4346e+06     91.7      NaN 
    3932    61.04    2606.5     10    2.2797e+06    91.07      NaN 
    3322    61.04    2735.6     10    2.3297e+06    91.84      2.5 
    3974    61.58    2899.2    9.5    2.3265e+06    91.59      1.5 
    5140    63.76    2815.8    9.5     2.407e+06    92.09        1 
    4074    65.93    2843.7    9.5    2.3097e+06     92.1        1 
    3571    66.67    2766.2    9.5    2.3867e+06    93.32        1 
    3737    67.75    2749.8    9.5    2.3796e+06    93.17        1 
    5087    69.95    2867.7    9.5    2.4641e+06    93.95        1 

[Walter Roberson]

My speculation is that dbload() is having problems because the first field is not quoted with "" but still contains characters ('Q')

If so then possibly the 'preprocess' option of dbload() would be useful. 'proprocess' takes a function handle, with the function accepting a character vector and returning a character vector. You could put in a function handle to a function that puts " quotes around the first portion of text, such as

d = dbload('data.csv', 'preprocess', @PrePro)

function S = PrePro(S)
  S = regexprep(S, '^(\w+)', '"$1"', 'lineanchors');
end