Fix tests that clashed with new numpy printing

Author: kkumer

.coverage

@@ -171,7 +171,7 @@ This is now a complete object and CFFs can be evaluated for some kinematics:
 .. code-block:: python
 
    >>> cff.ReH(g.DataPoint(xB=0.1, t=-0.3, Q2=6))
-   13.44851
+   np.float64(13.448510742222517)
 
 
 TFFs

docs/source/gpds.rst

@@ -93,7 +93,7 @@ User can also calculate other observables, like beam charge asymmetry
 .. code-block:: python
 
    >>> th_KM15.AC(pt)
-   0.138049
+   np.float64(0.1380493081710745)
 
 All implemented observables are listed :ref:`here<tab-observables>`.
 
@@ -104,7 +104,7 @@ example :math:`\mathfrak{Im}\mathcal{H}`
 .. code-block:: python
 
    >>> th_KM15.ImH(pt)
-   2.80754427
+   np.float64(2.807544271408012)
 
 
 .. note::
@@ -119,6 +119,6 @@ example :math:`\mathfrak{Im}\mathcal{H}`
 
    >>> pt = g.DataPoint(xB=0.348, t=-0.3, Q2=3)
    >>> th_KM15.ImH(pt)  # This will work
-   2.8075
+   np.float64(2.807544271408012)
 
 

docs/source/quickstart.rst

@@ -51,12 +51,12 @@ def beta(p: int, nf: int) -> float:
         beta = - B00 - B01 * nf
     elif p == 1:
         beta = - B10 - B11 * nf
-    elif p == 2:
-        beta = - 1428.50 + 279.611 * nf - 6.01852 * nf**2
-    elif p == 3:
-        beta = - 29243.0 + 6946.30 * nf - 405.089 * nf**2 - 1.49931 * nf**3
+#     elif p == 2:
+#         beta = - 1428.50 + 279.611 * nf - 6.01852 * nf**2
+#     elif p == 3:
+#         beta = - 29243.0 + 6946.30 * nf - 405.089 * nf**2 - 1.49931 * nf**3
     else:
-        raise ValueError('NNNNLO not yet implemented :-)')
+        raise ValueError('Only LO and NLO implemented!')
 
     return beta
 

src/gepard/qcd.py

@@ -6,7 +6,7 @@
 from typing import Tuple
 
 import numpy as np
-from scipy.special.orthogonal import p_roots
+from scipy.special import p_roots
 
 roots4, weights4 = p_roots(4)  # roots and weigths for 4-th order quadrature
 

src/gepard/quadrature.py

@@ -144,9 +144,9 @@ def predict(self, pt: data.DataPoint, uncertainty: bool = False, **kwargs) -> fl
                     var += (dfdp[p]*self.parameters_errors[p])**2
             else:
                 print('Theory has neither covariance matrix, nor parameters_errors.')
-            result = (fun(pt), sqrt(var))
+            result = (float(fun(pt)), float(sqrt(var)))
         else:
-            result = fun(pt)
+            result = float(fun(pt))
 
         if 'parameters' in kwargs:
             # restore old values