FORM should note lack of copula in human-readable error

[zdelrosario]

FORM routines should provide a human-readable error when a model without copula is provided.

Reprex follows:

import grama as gr

md = (
    gr.Model()
    >> gr.cp_vec_function(
        fun=lambda df: gr.df_make(g=df.x - 1),
        var=["x"],
        out=["g"],
    )
    >> gr.cp_marginals(x=dict(dist="norm", loc=0, scale=1))
)

(
    md
    >> gr.ev_form_pma(df_det="nom", betas=dict(g=3))
)
#> Traceback (most recent call last):
#> <ipython-input-5-ada1eb74ad57> in <module>
#>       1 (
#> ----> 2     md
#>       3     >> gr.ev_form_pma(df_det="nom", betas=dict(g=3))
#>       4 )
#> ~/Git/py_grama/grama/tools.py in __rrshift__(self, other)
#>     286                 other_copy = copy_meta(other, other_copy)
#>     287 
#> --> 288         result = self.function(other_copy)
#>     289 
#>     290         for p in self.chained_pipes:
#> ~/Git/py_grama/grama/tools.py in <lambda>(x)
#>     301 
#>     302     def __call__(self, *args, **kwargs):
#> --> 303         return pipe(lambda x: self.function(x, *args, **kwargs))
#>     304 
#>     305 
#> ~/Bin/anaconda3/lib/python3.8/site-packages/toolz/functoolz.py in __call__(self, *args, **kwargs)
#>     301     def __call__(self, *args, **kwargs):
#>     302         try:
#> --> 303             return self._partial(*args, **kwargs)
#>     304         except TypeError as exc:
#>     305             if self._should_curry(args, kwargs, exc):
#> ~/Git/py_grama/grama/eval_tail.py in eval_form_pma(model, betas, cons, df_corr, df_det, append, tol, n_maxiter, n_restart, verbose)
#>     134             res_all = []
#>     135             for jnd in range(n_restart):
#> --> 136                 res = minimize(
#>     137                     objective,
#>     138                     z0,
#> ~/.local/lib/python3.8/site-packages/scipy/optimize/_minimize.py in minimize(fun, x0, args, method, jac, hess, hessp, bounds, constraints, tol, callback, options)
#>     625         return _minimize_cobyla(fun, x0, args, constraints, **options)
#>     626     elif meth == 'slsqp':
#> --> 627         return _minimize_slsqp(fun, x0, args, jac, bounds,
#>     628                                constraints, callback=callback, **options)
#>     629     elif meth == 'trust-constr':
#> ~/.local/lib/python3.8/site-packages/scipy/optimize/slsqp.py in _minimize_slsqp(func, x0, args, jac, bounds, constraints, maxiter, ftol, iprint, disp, eps, callback, finite_diff_rel_step, **unknown_options)
#>     373 
#>     374     # ScalarFunction provides function and gradient evaluation
#> --> 375     sf = _prepare_scalar_function(func, x, jac=jac, args=args, epsilon=eps,
#>     376                                   finite_diff_rel_step=finite_diff_rel_step,
#>     377                                   bounds=new_bounds)
#> ~/.local/lib/python3.8/site-packages/scipy/optimize/optimize.py in _prepare_scalar_function(fun, x0, jac, args, bounds, epsilon, finite_diff_rel_step, hess)
#>     259     # ScalarFunction caches. Reuse of fun(x) during grad
#>     260     # calculation reduces overall function evaluations.
#> --> 261     sf = ScalarFunction(fun, x0, args, grad, hess,
#>     262                         finite_diff_rel_step, bounds, epsilon=epsilon)
#>     263 
#> ~/.local/lib/python3.8/site-packages/scipy/optimize/_differentiable_functions.py in __init__(self, fun, x0, args, grad, hess, finite_diff_rel_step, finite_diff_bounds, epsilon)
#>     134 
#>     135         self._update_fun_impl = update_fun
#> --> 136         self._update_fun()
#>     137 
#>     138         # Gradient evaluation
#> ~/.local/lib/python3.8/site-packages/scipy/optimize/_differentiable_functions.py in _update_fun(self)
#>     224     def _update_fun(self):
#>     225         if not self.f_updated:
#> --> 226             self._update_fun_impl()
#>     227             self.f_updated = True
#>     228 
#> ~/.local/lib/python3.8/site-packages/scipy/optimize/_differentiable_functions.py in update_fun()
#>     131 
#>     132         def update_fun():
#> --> 133             self.f = fun_wrapped(self.x)
#>     134 
#>     135         self._update_fun_impl = update_fun
#> ~/.local/lib/python3.8/site-packages/scipy/optimize/_differentiable_functions.py in fun_wrapped(x)
#>     128         def fun_wrapped(x):
#>     129             self.nfev += 1
#> --> 130             return fun(x, *args)
#>     131 
#>     132         def update_fun():
#> ~/Git/py_grama/grama/eval_tail.py in objective(z)
#>     112                 ## Transform: standard normal-to-random variable
#>     113                 df_norm = DataFrame(data=[z], columns=model.var_rand)
#> --> 114                 df_rand = model.norm2rand(df_norm)
#>     115                 df = model.var_outer(df_rand, df_det=df_inner)
#>     116 
#> ~/Git/py_grama/grama/core.py in norm2rand(self, df)
#>    1246         data = zeros((df.shape[0], self.n_var_rand))
#>    1247         for i in range(df.shape[0]):
#> -> 1248             data[i] = self.z2x(df[self.var_rand].iloc[i].values)
#>    1249 
#>    1250         return DataFrame(data=data, columns=self.var_rand)
#> ~/Git/py_grama/grama/core.py in z2x(self, z)
#>    1171         """
#>    1172         ## Correlate and map to uniform
#> -> 1173         u = self.density.copula.z2u(z)
#>    1174         ## Transform per marginal
#>    1175         x = zeros(self.n_var_rand)
#> AttributeError: 'NoneType' object has no attribute 'z2u'

^{Created on 2021-04-22 by the reprexpy package}