Problem interfacing matlab with python

Hi everybody, I'm calling some matlab functions into my python scripts and I'm having some problems, also because I'm quite new in MATLAB.

I generate a python package with matlab and I want to use it, instead of the matlab engine (is it a good strategy or should I prefer the engine?) By the way my question is: during the execution of my python script this exception is thrown "Operator '-' is not supported for operands of type 'cell'.", but if I run the entire script this not happens and I think that it's due to some type conversion, but I'm not sure how to correctly pass the arguments to the function. According to the error message I think that the exception is thrown when it performs the "S0=N-I0-Q0;" operation in the script error_SIR.

My python code is

import NetworkModelNationalIdentification as NMNI
mat_code = NMNI.initialize()
parameters,y,If = mat_code.id_and_sim(tab_data.tolist(),t_init,data_sep,initial_guess,population,total_active.tolist(),nargout=3)

note that:

• tab_data is a numpy array
• t_init is an integer value
• data_set is an integer value
• population is an integer value
• total_active is a numpy array
• intial_guess is a struct that I've defined as a python dictionary in this way:

initial_guess = {"It0": 100.0, "v": 0.97, "g": 1/14, "tau": 0.066}

and my MATLAB functions are:

function [pars,y,If] = id_and_sim(tab_data,ti,te,initial_guess,N,total_active)

%Take data needed for identification
data=tab_data(:,ti:te);
g=initial_guess.g;
It0=initial_guess.It0;
v=initial_guess.v;
tau=initial_guess.tau;

%Define the constrints for the identification
lim_sup=[v*10, tau*1.5, It0*1.3];
lim_inf=[v*0, tau*0.9, It0*0.7];


times=ti:te;

%identify the model parameters
pars=Identify_Model(data,initial_guess,lim_inf,lim_sup,times,N,total_active);
[v,tau,g,I0]=deal(pars{:});
clc;


%Simulate the model 
Q0=tab_data(1,ti);
S0=N-I0-Q0;
[t,y]=ode45(@(t,x) SIR(t,x,v,tau,g,total_active),times,[S0;I0;Q0]);
if length(times)==2
    t=[t(1),t(end)];
    y=[y(1,:);y(end,:)];
end
%Save the outputs in the form needed for the next steps
If=y(end-1,2);

y=reshape(y(:,3)',[size(y(:,3),1) 1]);

end

%%function called from the above one
function [pars] = Identify_Model(data,initial_guess,lim_inf,lim_sup,times,N,total_active)

scalefac=100;

%Initialize the initial guess for each parameter
v=initial_guess.v;
It0=initial_guess.It0;
tau=initial_guess.tau;
g=0.07;

lim_inf(3)=lim_inf(3)/scalefac;
lim_sup(3)=lim_sup(3)/scalefac;

%Identify the model parameters
options=optimset('MaxIter',100,'TolFun',1e-6,'TolX',1e-6);
parmin=fmincon(@(pars) error_SIR(data,[pars(1),pars(2),g,scalefac*pars(3)],times,N,total_active),[v,tau,It0],[],[],[],[],lim_inf,lim_sup,[],options);
pars=num2cell([parmin(1:2),scalefac*parmin(3)]);

pars=[pars(1),pars(2),g,pars(3)];

end

%function called from the above one
function [costo]=error_SIR(data,pars,tempi,N,totale_attivi)

%Assign the parameters
pars=num2cell(pars);
[v,tau,g,I0]=pars{:};
%Simulate the model
T=size(data,2);
Q0=data(1,1);
S0=N-I0-Q0;
[~,y]=ode45(@(t,x) SIR(t,x,v,tau,g,totale_attivi),tempi,[S0;I0;Q0]);
if length(tempi)==2
    y=[y(1,:);y(end,:)];
end

%Identify on the normalized data (Data/mean data)
costo=sum(((diff(sum(data,1))./sum(data(:,2:end),1))-(diff(sum(y(:,3),2))./sum(y(2:end,3),2))').^2);%+0.05*(pars{1}+pars{2})^2;

end