I'm trying to import a data set into Matlab without success. See attached file.

It appears that the issue is with the scientific notation. I've attempted to use fscanf(), dlmread() and importdata(). It either imports an empty set or I receive an error message. The following is my attempted code:

% Mike Beeman
% Peakfit Electrochemistry Pathogen Meat Detection 

clc
clear all
close all

%%Import Data
% fid = fopen('/Users/mbeeman6/Desktop/Bacteria Detection Research - Gale/Data/june 12th Crypto datatofidgetwithonmatlab.csv')
% 
% C = fscanf(fid,'%e') %,'HeaderLines',6);
% fclose all

data = dlmread('/Users/mbeeman6/Desktop/Bacteria Detection Research - Gale/Data/june 12th Crypto datatofidgetwithonmatlab.csv')

Thanks in advanced for any help you can offer.

Answer 1

[per isakson]

One way

ffs = 'june 12th Crypto datatofidgetwithonmatlab.csv';
fid = fopen( ffs, 'r' );
fmt = [ '"', repmat('%f',1,60), '"' ]; % Include the '"'
cac = textscan( fid, fmt, 'Headerlines',5, 'CollectOutput',true, 'Delimiter',',' );
fclose(fid);

and inspect the result

>> v = cac{1};
>> whos v
  Name        Size            Bytes  Class     Attributes
  v         170x60            81600  double

[Mike Beeman]

Thanks for the help! It's much appreciated Isakson!

Answer 2

[Walter Roberson]

dlmread() cannot handle quoted strings. It also cannot text headers unless you are using roughly R2014a onwards and tell it to skip those lines.

as_string = fileread('june 12th Crypto datatofidgetwithonmatlab.csv');
as_string(as_string == '"') = [];
as_lines = regexp(as_string, '\r?\n', 'split');
column_headers = regexp(as_lines{4}, ',', 'split');
data_units = regexp(as_lines{5}, ',', 'split');
if length(column_headers) < length(data_units); column_headers(length(data_units)) = {''}; end    %last column of headers might be missing
data_items = regexp(as_lines(6:end), ',', 'split');
data_cell = vertcat(data_items{:});
data_numeric = str2double(data_cell);

The key outputs here are column_headers (cell array of char vector), data_units (cell array of char vector), and data_numeric (numeric array).

Potentially you might want to go from there to

t = array2table(data_numeric, 'VariableNames', matlab.lang.makeUniqueStrings(matlab.lang.makeValidName(column_headers)));
t.Properties.VariableDescriptions = column_headers;
t.Properties.VariableUnits = data_units;

The Variable Names will not be the nicest, as they will come out as a mix of names like H3CryptoCavitationTest_SWV and names like x_17 . The x_* names are from columns whose headers were empty.

The variable descriptions and units will only be visible if you ask for the properties, or if you use summary(t)

[Mike Beeman]

Thanks for the help! It's much appreciated Walter Roberson!