Converting a Python code that deals with arrays to MATLAB

Hello Everyone,

I'M working on a spike sotring code in python but When I tried to convert it to MATLAB, the code did not work as expected. please can anyone help me to solve my issue ?

Python Code:

if np.max(tmpW) < max_thresh:
                tmpS = np.argmax(tmpW) +i

                tmpW = data[tmpS-(sw-offset):tmpS+(sw+offset)]

                ss = np.append(ss, tmpS)
                wf = np.append(wf, tmpW.reshape(1, sw*2), axis=0)

    # Remove duplicates
    ind = np.where(np.diff(ss) > 1)[0]
    ss = ss[ind]
    wf = wf[ind]

Matlab code:

if max(wave_form1)< max_thresh
        tmp_samp = find(max(wave_form1))+i
        wave_form1 = data(tmp_samp-(spike_window-offset):tmp_samp+(spike_window+offset))
        spike_samp=[spike_samp tmp_samp]
        wave_from = [wave_form wave_form1]
    end 

[Walter Roberson]

I would tend to suspect

        wave_from = [wave_form; wave_form1]

because of the axis=0

[Anas Wheba]

I tried but nothing worked, pleae can you help me ?. Im trying to align my spikes

[Walter Roberson]

You have

        wave_from = [wave_form wave_form1]

should probably be

       wave_form = [wave_form; wave_form1]

Notice the change of name of the destination variable, wave_from vs wave_form

[Anas Wheba]

yes it got fixed, but its not storing any data because of this line

wave_form1 = data(tmp_samp-(spike_window-offset):tmp_samp+(spike_window+offset));

orginal code in python:

tmpW = data[tmpS-(sw-offset):tmpS+(sw+offset)]       

[Walter Roberson]

At the time of the error, what is size(data), and the values tmp_samp, spike_window, and offset ? Also, what are the maximum and minimum values for tmp_samp as you iterate through the loop ?

[Yongjian Feng]

Also this line is suspicious:

tmp_samp = find(max(wave_form1))+i

Do you want this instead?

tmp_samp = find(wave_form1 == max(wave_form1))+i

Answer 1

[Vandit]

Hello,

After assuming that 'wave_form', 'wave_form1', 'spike_samp', 'wave_from', 'max_thresh', 'data, spike_window',' offset', and 'i' are defined correctly in your MATLAB environment, the issue in the MATLAB code lies in the lines:

tmp_samp = find(max(wave_form1)) + i; %Incorrect code
tmp_samp = find(wave_form1 == max(wave_form1)) + i; %Modified code

The reason for changing the above code is to find the indices of the maximum value(s) in 'wave_form1' rather than just the first occurrence. In the original Python code, 'np.argmax(tmpW)' returns the index of the first occurrence of the maximum value in 'tmpW'. However, in MATLAB, using 'find(max(wave_form1))' would only return the index of the first occurrence of the maximum value in 'wave_form1', which may not be the desired behavior.

wave_from = [wave_form wave_form1]; %Incorrect code
wave_form = [wave_form; wave_form1]; %Modified code

The reason for changing the code is to correctly append the 'wave_form1' waveform as a new row to the 'wave_form' matrix. In the original Python code, 'wf' is an array where each row represents a waveform. In the MATLAB code, the equivalent variable is 'wave_form'. To correctly append 'wave_form1' as a new row to 'wave_form', we need to use the semicolon (;) to separate the rows.

Hope this helps.

Thankyou