How to create the cvs file

[pingguoshala]

   Hi, I am a beginer of the Micro ML ,Your project is super cool, I like it very mauch! 
   In th pratical of " Handwritten digit classification with Arduino and MicroML", I'am trying to create csv file order :
              # put your samples in the dataset folder
              # one class per file
              # one feature vector per line, in CSV format
  But   it  aways  with a problem when Train and export the SVM classifier,I hope you offer exzample

csv files for me to study. Thank!

[eloquentarduino]

I attached the CSV I generated to run the example in the post. There's also a snippet to plot those samples, to get you an idea of what they should look like. Each line represents a feature vector. Cheers Simone Il giorno mer 26 feb 2020 alle ore 12:32 pingguoshala < notifications@github.com> ha scritto:

…

Hi, I am a beginer of the Micro ML ,Your project is super cool, I like it very mauch! In th pratical of " Handwritten digit classification with Arduino and MicroML", I'am trying to create csv file order : # put your samples in the dataset folder # one class per file # one feature vector per line, in CSV format But it aways with a problem when Train and export the SVM classifier,I hope you offer exzample csv files for me to study. Thank! — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#3?email_source=notifications&email_token=AOA7GNOLBPTCMEAQB4ZCXX3REZHMTA5CNFSM4K4C4KLKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IQM35NQ>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AOA7GNP4M223JULQ3LJPEI3REZHMTANCNFSM4K4C4KLA> .

[pingguoshala]

     Thank to your reply, I copy the feature vector line  from Arduino Serial monitor  directly to the opend txt file, and press Enter key to  the next line, do the same  Until the last feature vector  line ,then I change the .txt file to the .csv file.and put it to the dataset folder. Is that right ？
    when I with load_features() funtion ,I meet the error 
   Traceback (most recent call last):

File "D:/1.py", line 16, in
features, classmap = load_features('data')
File "D:/1.py", line 11, in load_features
samples = np.loadtxt(filename)
File "C:\Users\Administrator\AppData\Local\Programs\Python\Python36\lib\site-packages\numpy\lib\npyio.py", line 1159, in loadtxt
for x in read_data(_loadtxt_chunksize):
File "C:\Users\Administrator\AppData\Local\Programs\Python\Python36\lib\site-packages\numpy\lib\npyio.py", line 1087, in read_data
items = [conv(val) for (conv, val) in zip(converters, vals)]
File "C:\Users\Administrator\AppData\Local\Programs\Python\Python36\lib\site-packages\numpy\lib\npyio.py", line 1087, in
items = [conv(val) for (conv, val) in zip(converters, vals)]
File "C:\Users\Administrator\AppData\Local\Programs\Python\Python36\lib\site-packages\numpy\lib\npyio.py", line 794, in floatconv
return float(x)
ValueError: could not convert string to float: '0.00,1.00,1.00,1.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.

[eloquentarduino]

Another guy had the same problem. He told me he solved replacing np.loadtxt(filename) with np.genfromtxt(filename, delimiter=',', skip_header=1) Let me know if it works Il giorno mer 26 feb 2020 alle ore 14:39 pingguoshala < notifications@github.com> ha scritto:

…

Thank to your reply, I copy the feature vector line from Arduino Serial monitor directly to the opend txt file, and press Enter key to the next line, do the same Until the last feature vector line ,then I change the .txt file to the .csv file.and put it to the dataset folder. Is that right ？ when I with load_features() funtion ,I meet the error Traceback (most recent call last): File "D:/1.py", line 16, in features, classmap = load_features('data') File "D:/1.py", line 11, in load_features samples = np.loadtxt(filename) File "C:\Users\Administrator\AppData\Local\Programs\Python\Python36\lib\site-packages\numpy\lib\npyio.py", line 1159, in loadtxt for x in read_data(_loadtxt_chunksize): File "C:\Users\Administrator\AppData\Local\Programs\Python\Python36\lib\site-packages\numpy\lib\npyio.py", line 1087, in read_data items = [conv(val) for (conv, val) in zip(converters, vals)] File "C:\Users\Administrator\AppData\Local\Programs\Python\Python36\lib\site-packages\numpy\lib\npyio.py", line 1087, in items = [conv(val) for (conv, val) in zip(converters, vals)] File "C:\Users\Administrator\AppData\Local\Programs\Python\Python36\lib\site-packages\numpy\lib\npyio.py", line 794, in floatconv return float(x) ValueError: could not convert string to float: '0.00,1.00,1.00,1.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#3?email_source=notifications&email_token=AOA7GNOINS3GMGXBV5Y5OLTREZWH5A5CNFSM4K4C4KLKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOENAIFSA#issuecomment-591430344>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AOA7GNO2REMOBPBYAANNCKTREZWH5ANCNFSM4K4C4KLA> .

[pingguoshala]

Thanks ! I solved replacing

np.loadtxt(filename)

with

samples = np.loadtxt(filename,dtype=float, delimiter=',')

It works well.

I replacing np.loadtxt(filename) with

np.genfromtxt(filename, delimiter=',')

It works well.

[eloquentarduino]

Good to know, I'll update my post Il giorno gio 27 feb 2020 alle ore 02:32 pingguoshala < notifications@github.com> ha scritto:

…

Thanks ! I solved replacing np.loadtxt(filename) with samples = np.loadtxt(filename,dtype=float, delimiter=',') It works well. But I replacing np.loadtxt(filename) with np.genfromtxt(filename, delimiter=',', skip_header=1) An error occurred Traceback (most recent call last): File "D:\1.py", line 24, in features, classmap = load_features('data') File "D:\1.py", line 21, in load_features samples = np.hstack((samples, labels)) File "<*array_function* internals>", line 6, in hstack File "C:\Users\Administrator\AppData\Local\Programs\Python\Python36\lib\site-packages\numpy\core\shape_base.py", line 343, in hstack return _nx.concatenate(arrs, 0) File "<*array_function* internals>", line 6, in concatenate ValueError: all the input arrays must have same number of dimensions, but the array at index 0 has 1 dimension(s) and the array at index 1 has 2 dimension(s) — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#3?email_source=notifications&email_token=AOA7GNKZR4SLAYVVZQFS253RE4JZXA5CNFSM4K4C4KLKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOENCRVBQ#issuecomment-591731334>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AOA7GNL4DVEUKH3BYYNM33DRE4JZXANCNFSM4K4C4KLA> .