kcorrect Calculating k-correct in kcorrect.kcorrect.kcorrectGST

Hello! There's a issue when I use the Galex-SDSS-2MASS script. The error message is "coeffs must be 2-dimensional if redshift is 1-dimensional" or others errors like this one. When I use the same script but using now kcorrect.kcorrect.kcorrectSDSS the code runs normaly.

Oct 19 '23 14:10 Raphaelgsousa

Can you be more specific about which script you are trying to run? and send the traceback?

Oct 19 '23 15:10 blanton144

This one runs normaly calculating k:

def calc_kcorrectionSDSS(redshift):
    responses = ['sdss_r0', 'sdss_g0', 'sdss_u0', 'sdss_i0', 'sdss_z0']
    kc = kcorrect.kcorrect.KcorrectSDSS(responses=responses, abcorrect = True)

    redshift = redshift
    maggies = [Amostrapequenateste['cModelFlux_r'],Amostrapequenateste['cModelFlux_g'],Amostrapequenateste['cModelFlux_u'],Amostrapequenateste['cModelFlux_i'],Amostrapequenateste['cModelFlux_z']]
    ivar = [Amostrapequenateste['cModelFluxIvar_r'],Amostrapequenateste['cModelFluxIvar_g'],Amostrapequenateste['cModelFluxIvar_u'],Amostrapequenateste['cModelFluxIvar_i'],Amostrapequenateste['cModelFluxIvar_z']]

    # "coeffs" is a [2,5]-array coefficients multiplying each template
    coeffs = kc.fit_coeffs(redshift=redshift, maggies=maggies,ivar=ivar)

    # "k" is a [2,4]-array with the K-corrections in magnitude units
    k = kc.kcorrect(redshift=redshift, coeffs=coeffs,band_shift=0.1)
    tabela_kcorr = pd.DataFrame(k, columns = ['K-corr (r)','K-corr (g)','K-corr (u)','K-corr (i)','K-corr (z)'])
    return print(k)

But this one, pretty much the same but now using GST and "galex_NUV" doesn't run:

responsesGST = ['galex_NUV','sdss_r0']
kcGST = kcorrect.kcorrect.KcorrectGST(responses=responsesGST, abcorrect = True)
lista_redshift = Amostrapequenateste['z']
redshiftGST = lista_redshift
maggiesGST = [Amostrapequenateste['nuv_flux (maggies)'].to_numpy(dtype=np.float32),Amostrapequenateste['cModelFlux_r'].to_numpy(dtype=np.float32)]
ivarGST = [Amostrapequenateste['NUV Ivar'].to_numpy(dtype=np.float32), Amostrapequenateste['cModelFluxIvar_r'].to_numpy(dtype=np.float32)]

# "coeffs" is a [2,5]-array coefficients multiplying each template
coeffsGST = kcGST.fit_coeffs(redshift=redshiftGST, maggies=maggiesGST,ivar=ivarGST)

# "k" is a [2,4]-array with the K-corrections in magnitude units
kGST = kcGST.kcorrect(redshift=redshiftGST, coeffs=coeffsGST,band_shift=0.1)
tabela_kcorrGST = pd.DataFrame(kGST, columns = ['K-corr (NUV)','K-corr (r)'])

With this error:

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Cell In[54], line 11
      8 coeffsGST = kcGST._fit_coeffs(redshift=redshiftGST, maggies=maggiesGST,ivar=ivarGST)
     10 # "k" is a [2,4]-array with the K-corrections in magnitude units
---> 11 kGST = kcGST.kcorrect(redshift=redshiftGST, coeffs=coeffsGST,band_shift=0.1)
     12 print(k)

File ~/anaconda3/lib/python3.10/site-packages/kcorrect/kcorrect.py:423, in Kcorrect.kcorrect(self, redshift, coeffs, band_shift)
    401     def kcorrect(self, redshift=None, coeffs=None, band_shift=0.):
    402         """Return K-correction in all bands
    403 
    404         Parameters
   (...)
    420             K-correction from input to output magnitudes
    421 """
    422         (array, n, redshift, d1,
--> 423          d2, coeffs) = self._process_inputs(redshift=redshift, maggies=None,
    424                                             ivar=None, coeffs=coeffs)
    426         # maggies associated with bandpass R at observed z
    427         maggies_in = self.reconstruct(redshift=redshift, coeffs=coeffs)

File ~/anaconda3/lib/python3.10/site-packages/kcorrect/fitter.py:314, in Fitter._process_inputs(self, redshift, maggies, ivar, coeffs)
    312 if(array):
    313     if(len(coeffs.shape) != 2):
--> 314         raise ValueError("coeffs must be 2-dimensional if redshift is 1-dimensional")
    315     if(coeffs.shape[0] != n):
    316         raise ValueError("ivar must have values for each redshift")

ValueError: coeffs must be 2-dimensional if redshift is 1-dimensional

Where "Amostrapequenateste" is the dataframe of my sample.

Oct 20 '23 15:10 Raphaelgsousa

Just a question first: You are using KcorrectGST but you are specifying the responses?

If you are specifying the responses, just use the base Kcorrect class.

On Fri, Oct 20, 2023 at 11:21 AM Raphael Gomes Sousa < @.***> wrote:

This one runs normaly calculating k:

def calc_kcorrectionSDSS(redshift): responses = ['sdss_r0', 'sdss_g0', 'sdss_u0', 'sdss_i0', 'sdss_z0'] kc = kcorrect.kcorrect.KcorrectSDSS(responses=responses, abcorrect = True)
redshift = redshift
maggies = [Amostrapequenateste['cModelFlux_r'],Amostrapequenateste['cModelFlux_g'],Amostrapequenateste['cModelFlux_u'],Amostrapequenateste['cModelFlux_i'],Amostrapequenateste['cModelFlux_z']]
ivar = [Amostrapequenateste['cModelFluxIvar_r'],Amostrapequenateste['cModelFluxIvar_g'],Amostrapequenateste['cModelFluxIvar_u'],Amostrapequenateste['cModelFluxIvar_i'],Amostrapequenateste['cModelFluxIvar_z']]

# "coeffs" is a [2,5]-array coefficients multiplying each template
coeffs = kc.fit_coeffs(redshift=redshift, maggies=maggies,ivar=ivar)

# "k" is a [2,4]-array with the K-corrections in magnitude units
k = kc.kcorrect(redshift=redshift, coeffs=coeffs,band_shift=0.1)
tabela_kcorr = pd.DataFrame(k, columns = ['K-corr (r)','K-corr (g)','K-corr (u)','K-corr (i)','K-corr (z)'])
return print(k)
But this one, pretty much the same but now using GST and "galex_NUV" doesn't run:

responsesGST = ['galex_NUV','sdss_r0']kcGST = kcorrect.kcorrect.KcorrectGST(responses=responsesGST, abcorrect = True)lista_redshift = Amostrapequenateste['z']redshiftGST = lista_redshiftmaggiesGST = [Amostrapequenateste['nuv_flux (maggies)'].to_numpy(dtype=np.float32),Amostrapequenateste['cModelFlux_r'].to_numpy(dtype=np.float32)]ivarGST = [Amostrapequenateste['NUV Ivar'].to_numpy(dtype=np.float32), Amostrapequenateste['cModelFluxIvar_r'].to_numpy(dtype=np.float32)]

"coeffs" is a [2,5]-array coefficients multiplying each templatecoeffsGST = kcGST.fit_coeffs(redshift=redshiftGST, maggies=maggiesGST,ivar=ivarGST)

"k" is a [2,4]-array with the K-corrections in magnitude unitskGST = kcGST.kcorrect(redshift=redshiftGST, coeffs=coeffsGST,band_shift=0.1)tabela_kcorrGST = pd.DataFrame(kGST, columns = ['K-corr (NUV)','K-corr (r)'])

With this error:

---------------------------------------------------------------------------ValueError Traceback (most recent call last)Cell In[54], line 11 8 coeffsGST = kcGST._fit_coeffs(redshift=redshiftGST, maggies=maggiesGST,ivar=ivarGST) 10 # "k" is a [2,4]-array with the K-corrections in magnitude units---> 11 kGST = kcGST.kcorrect(redshift=redshiftGST, coeffs=coeffsGST,band_shift=0.1) 12 print(k) File ~/anaconda3/lib/python3.10/site-packages/kcorrect/kcorrect.py:423, in Kcorrect.kcorrect(self, redshift, coeffs, band_shift) 401 def kcorrect(self, redshift=None, coeffs=None, band_shift=0.): 402 """Return K-correction in all bands 403 404 Parameters (...) 420 K-correction from input to output magnitudes 421 """ 422 (array, n, redshift, d1,--> 423 d2, coeffs) = self._process_inputs(redshift=redshift, maggies=None, 424 ivar=None, coeffs=coeffs) 426 # maggies associated with bandpass R at observed z 427 maggies_in = self.reconstruct(redshift=redshift, coeffs=coeffs) File ~/anaconda3/lib/python3.10/site-packages/kcorrect/fitter.py:314, in Fitter._process_inputs(self, redshift, maggies, ivar, coeffs) 312 if(array): 313 if(len(coeffs.shape) != 2):--> 314 raise ValueError("coeffs must be 2-dimensional if redshift is 1-dimensional") 315 if(coeffs.shape[0] != n): 316 raise ValueError("ivar must have values for each redshift") ValueError: coeffs must be 2-dimensional if redshift is 1-dimensional

Where "Amostrapequenateste" is the dataframe of my sample.

— Reply to this email directly, view it on GitHub https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_blanton144_kcorrect_issues_5-23issuecomment-2D1772940636&d=DwMFaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=2nqg0fcNitaQTtJ4c9MSjg&m=ynBkDWFpderlldQVgRw-3q4OrTL0-sj0u44-xDWEy__BBFmQySsuBc7hzp7roiS9&s=mc58aCovZBbC9ezubl6wqwz_88HWBOIfTqZNzgvxWVw&e=, or unsubscribe https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_AAODDFQV2RR5W2D2ADQSHVDYAKJHLAVCNFSM6AAAAAA6HJA5WKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONZSHE2DANRTGY&d=DwMFaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=2nqg0fcNitaQTtJ4c9MSjg&m=ynBkDWFpderlldQVgRw-3q4OrTL0-sj0u44-xDWEy__BBFmQySsuBc7hzp7roiS9&s=hc1N66fRI2jQPyoWe5DGnHt9BzmpInXkDmQaas2Je04&e= . You are receiving this because you commented.Message ID: @.***>

-- Michael Blanton, Professor Department of Physics New York University

Oct 20 '23 17:10 blanton144

Also, can you verify that the behavior is really different in the two cases? You are giving "redshift" in one case, whose nature I can't tell, and "lista_redshift" in the other. Try making sure you are passing EXACTLY the same structures in.

In particular, look at the shape of the redshift, maggies, and ivar arrays for fit_coeffs() and coeffs arrays you are sending the kcorrect() method call. If you can just send me those for the two cases I could try to figure it out.

Oct 20 '23 18:10 blanton144