Not compatible with pandas >= 0.19.0

[pgkirsch]

fit.py uses pandas.Series.reshape, which, per the pandas docs has been deprecated since pandas 0.19.0:

https://pandas.pydata.org/pandas-docs/version/0.22/generated/pandas.Series.reshape.html

[bqpd]

oh wild, where does it use it? probably the easiest fix is to convert to numpy before the reshape.

[whoburg]

Thanks @pgkirsch. Do you have a simple test or example that produces this issue?

[pgkirsch]

I was mistaken. fit.py doesn't use pandas.Series.reshape; this occurs when I use a pd.Series object as my input.

Example is if we make one of the input arrays for gpfit/examples/ex63.py a pd.Series:

from numpy import log, exp, log10, vstack                                        
from gpfit.fit import fit                                                        
from numpy.random import random_sample                                           
import pandas as pd                                                                                                                                 
Vdd = pd.Series(random_sample(1000,) + 1)                                        
Vth = 0.2*random_sample(1000,) + 0.2                                                                                                          
P = Vdd**2 + 30*Vdd*exp(-(Vth-0.06*Vdd)/0.039)                                   
u = vstack((Vdd,Vth))                                                            
x = log(u)                                                                       
y = log(P)                                                                       
K = 4                                                                                                                                                       
cstrt, rmsErr = fit(x, y, K, "ISMA")

yields

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
~/Optimization/gpfit/gpfit/examples/pd.py in <module>
     12 K = 4
     13 
---> 14 cstrt, rmsErr = fit(x,y,K,"ISMA")

~/Optimization/gpfit/gpfit/fit.py in fit(xdata, ydata, K, ftype)
     78             fitdata["ub%d" % i] = exp(max(xdata.T[i]))
     79 
---> 80     params = get_params(ftype, K, xdata, ydata)
     81 
     82     # A: exponent parameters, B: coefficient parameters

~/Optimization/gpfit/gpfit/fit.py in get_params(ftype, K, xdata, ydata)
     23         return r, drdp
     24 
---> 25     ba = ba_init(xdata, ydata.reshape(ydata.size, 1), K).flatten('F')
     26 
     27     if ftype == "ISMA":

/opt/anaconda3/lib/python3.8/site-packages/pandas/core/generic.py in __getattr__(self, name)
   5137             if self._info_axis._can_hold_identifiers_and_holds_name(name):
   5138                 return self[name]
-> 5139             return object.__getattribute__(self, name)
   5140 
   5141     def __setattr__(self, name: str, value) -> None:

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

I agree with @bqpd's suggestion to make sure all input arrays are numpy arrays as a pre-process step.

[pgkirsch]

Or make it a ux change and raise an error if the input type isn't a numpy array.