AttributeError: 'Series' object has no attribute 'real' using EIS_plot method

[swsyoon]

I was able to fit an experimental data using ex.EIS_fit(params=params, circuit='R-RQ-RQ',weight_func='modulus')

However when I tried to plot the data using ex2.EIS_plot(legend='potential', bode='log', fitting='on'), it generates the error.

  File "~\Miniconda3\envs\impedance\lib\site-packages\PyEIS\PyEIS.py", line 4701, in EIS_plot
    ax.plot(self.circuit_fit[i].real, -self.circuit_fit[i].imag, lw=0, marker='o', ms=8, mec='r', mew=1, mfc='none', label='')

  File "~\Miniconda3\envs\impedance\lib\site-packages\pandas\core\generic.py", line 5136, in __getattr__
    return object.__getattribute__(self, name)

AttributeError: 'Series' object has no attribute 'real'

I am using python 3.7.9 in windows 10.

[mfjimenezmoreno]

Having the same problem when I follow the anexed tutorial (Doing this on Python 3.9.2).

I guess it has to do that the column is named 're' instead of 'real'? Seems to me that is fitting correctly, but fails at the moment it tries to plot.

[gimdeokhwan]

It is same issue as an "Error using Lin_kk function"

I've fixed the this issue, but have no authority of "pull request"
but I've tested attached code, works well.

Reason why issue happened is self.KK_circuit_fit[i].real and self.KK_circuit_fit[i].imag not aligned to array data format.

so it should be changed following
(I'm in office so not possible to upload the screen shot and code)

       #changed by Gim
        length_of_kk_circuit_fit = len(self.KK_circuit_fit[i])
        KK_circuit_fit_real = []
        KK_circuit_fit_imag = []
        for idx in range(length_of_kk_circuit_fit):
            KK_circuit_fit_real.append(self.KK_circuit_fit[i][idx+length_of_kk_circuit_fit].real)
            KK_circuit_fit_imag.append(self.KK_circuit_fit[i][idx + length_of_kk_circuit_fit].imag)

        KK_circuit_fit_real = np.array(KK_circuit_fit_real)
        KK_circuit_fit_imag = np.array(KK_circuit_fit_imag)
        self.KK_rr_re.append(residual_real(re=self.df[i].re, fit_re=KK_circuit_fit_real, fit_im=-KK_circuit_fit_imag))#relative residuals for the real part
        self.KK_rr_im.append(residual_imag(im=self.df[i].im, fit_re=KK_circuit_fit_real, fit_im=-KK_circuit_fit_imag)) #relative residuals for the imag part
        # changed by Gim
        
        #original source code
        #self.KK_rr_re.append(residual_real(re=self.df[i].re, fit_re=self.KK_circuit_fit[i].real, fit_im=-self.KK_circuit_fit[i].imag)) #relative residuals for the real part
        #self.KK_rr_im.append(residual_imag(im=self.df[i].im, fit_re=self.KK_circuit_fit[i].real, fit_im=-self.KK_circuit_fit[i].imag)) #relative residuals for the imag part

[FabianPascher]

One solution is to change the 4701 entry
from (original)
ax.plot(self.circuit_fit[i].real, -self.circuit_fit[i].imag, lw=0, marker='o', ms=8, mec='r', mew=1, mfc='none', label='')
to (my solution)
ax.plot(self.circuit_fit[i].values.real, -self.circuit_fit[i].values.imag, lw=0, marker='o', ms=8, mec='r', mew=1, mfc='none', label='')

[DouglasGuillen99]

One solution is to change the 4701 entry from (original) ax.plot(self.circuit_fit[i].real, -self.circuit_fit[i].imag, lw=0, marker='o', ms=8, mec='r', mew=1, mfc='none', label='') to (my solution) ax.plot(self.circuit_fit[i].values.real, -self.circuit_fit[i].values.imag, lw=0, marker='o', ms=8, mec='r', mew=1, mfc='none', label='')

@FabianPascher I want to apply your solution, I need to fit experimental data as well and I need to fix the same issue. Please reply if you are able to see my comment.

[DouglasGuillen99]

It is same issue as an "Error using Lin_kk function"

I've fixed the this issue, but have no authority of "pull request" but I've tested attached code, works well.

Reason why issue happened is self.KK_circuit_fit[i].real and self.KK_circuit_fit[i].imag not aligned to array data format.

so it should be changed following (I'm in office so not possible to upload the screen shot and code)

       #changed by Gim
        length_of_kk_circuit_fit = len(self.KK_circuit_fit[i])
        KK_circuit_fit_real = []
        KK_circuit_fit_imag = []
        for idx in range(length_of_kk_circuit_fit):
            KK_circuit_fit_real.append(self.KK_circuit_fit[i][idx+length_of_kk_circuit_fit].real)
            KK_circuit_fit_imag.append(self.KK_circuit_fit[i][idx + length_of_kk_circuit_fit].imag)

        KK_circuit_fit_real = np.array(KK_circuit_fit_real)
        KK_circuit_fit_imag = np.array(KK_circuit_fit_imag)
        self.KK_rr_re.append(residual_real(re=self.df[i].re, fit_re=KK_circuit_fit_real, fit_im=-KK_circuit_fit_imag))#relative residuals for the real part
        self.KK_rr_im.append(residual_imag(im=self.df[i].im, fit_re=KK_circuit_fit_real, fit_im=-KK_circuit_fit_imag)) #relative residuals for the imag part
        # changed by Gim
        
        #original source code
        #self.KK_rr_re.append(residual_real(re=self.df[i].re, fit_re=self.KK_circuit_fit[i].real, fit_im=-self.KK_circuit_fit[i].imag)) #relative residuals for the real part
        #self.KK_rr_im.append(residual_imag(im=self.df[i].im, fit_re=self.KK_circuit_fit[i].real, fit_im=-self.KK_circuit_fit[i].imag)) #relative residuals for the imag part

@gimdeokhwan I want to apply your solution, I need to fit experimental data as well and I need to fix the same issue. Please reply if you are able to see my comment.

[zleung9]

One solution is to change the 4701 entry from (original) ax.plot(self.circuit_fit[i].real, -self.circuit_fit[i].imag, lw=0, marker='o', ms=8, mec='r', mew=1, mfc='none', label='') to (my solution) ax.plot(self.circuit_fit[i].values.real, -self.circuit_fit[i].values.imag, lw=0, marker='o', ms=8, mec='r', mew=1, mfc='none', label='')

@FabianPascher I've applied your solution and it worked. Note that one needs to change self.circuit_fit[i] to self.circuit_fit[i].values for all occurances in this file.