Pass rcond=None to numpy.linalg.lstsq to silence warning

Silence numpy warnings and use new default behavior

From numpy documentation:

Changed in version 1.14.0: If not set, a FutureWarning is given.
The previous default of -1 will use the machine precision as rcond
parameter, the new default will use the machine precision times max(M, N).
To silence the warning and use the new default, use rcond=None,
to keep using the old behavior, use rcond=-1.

rmgpy/data/kinetics/groups.py

@@ -409,7 +409,7 @@ def generateGroupAdditivityValues(self, trainingSet, kunits, method='Arrhenius')
             b = numpy.array(b)
             kdata = numpy.array(kdata)
 
-            x, residues, rank, s = numpy.linalg.lstsq(A, b)
+            x, residues, rank, s = numpy.linalg.lstsq(A, b, rcond=None)
 
             for t, T in enumerate(Tdata):
 
@@ -513,7 +513,7 @@ def generateGroupAdditivityValues(self, trainingSet, kunits, method='Arrhenius')
             b = numpy.array(b)
             kdata = numpy.array(kdata)
 
-            x, residues, rank, s = numpy.linalg.lstsq(A, b)
+            x, residues, rank, s = numpy.linalg.lstsq(A, b, rcond=None)
 
             # Store the results
             self.top[0].data = Arrhenius(
@@ -564,7 +564,7 @@ def generateGroupAdditivityValues(self, trainingSet, kunits, method='Arrhenius')
             A = numpy.array(A)
             b = numpy.array(b)
 
-            x, residues, rank, s = numpy.linalg.lstsq(A, b)
+            x, residues, rank, s = numpy.linalg.lstsq(A, b, rcond=None)
 
             # Store the results
             self.top[0].data = Arrhenius(

rmgpy/kinetics/arrhenius.pyx

@@ -162,7 +162,7 @@ cdef class Arrhenius(KineticsModel):
             for n in range(b.size):
                 A[n,:] *= weights[n]
                 b[n] *= weights[n]
-        x, residues, rank, s = numpy.linalg.lstsq(A,b)
+        x, residues, rank, s = numpy.linalg.lstsq(A, b, rcond=None)
 
         # Determine covarianace matrix to obtain parameter uncertainties
         count = klist.size

rmgpy/kinetics/chebyshev.pyx

@@ -217,7 +217,7 @@ cdef class Chebyshev(PDepKineticsModel):
                 b[p1*nT+t1] = log10(K[t1,p1])
 
         # Do linear least-squares fit to get coefficients
-        x, residues, rank, s = numpy.linalg.lstsq(A, b)
+        x, residues, rank, s = numpy.linalg.lstsq(A, b, rcond=None)
 
         # Extract coefficients
         coeffs = numpy.zeros((degreeT,degreeP), numpy.float64)

rmgpy/kinetics/surface.pyx

@@ -157,7 +157,7 @@ cdef class StickingCoefficient(KineticsModel):
             for n in range(b.size):
                 A[n, :] *= weights[n]
                 b[n] *= weights[n]
-        x, residues, rank, s = numpy.linalg.lstsq(A, b)
+        x, residues, rank, s = numpy.linalg.lstsq(A, b, rcond=None)
 
         # Determine covarianace matrix to obtain parameter uncertainties
         count = klist.size

rmgpy/statmech/torsion.pyx

@@ -539,7 +539,7 @@ cdef class HinderedRotor(Torsion):
             # This row forces dV/dangle = 0 at angle = 0
             for m in range(numterms):
                 A[N,m+numterms] = 1
-            x, residues, rank, s = numpy.linalg.lstsq(A, b)
+            x, residues, rank, s = numpy.linalg.lstsq(A, b, rcond=None)
             fit = numpy.dot(A,x)
             x *= 0.001
             # This checks if there are any negative values in the forier fit.

rmgpy/thermo/wilhoit.pyx

@@ -277,7 +277,7 @@ cdef class Wilhoit(HeatCapacityModel):
                 for j in range(4):
                     A[i,j] = (y*y*y - y*y) * y**j
                 b[i] = ((Cpdata[i] - Cp0) / (CpInf - Cp0) - y*y)
-            x, residues, rank, s = numpy.linalg.lstsq(A, b)
+            x, residues, rank, s = numpy.linalg.lstsq(A, b, rcond=None)
 
             self.B = (float(B),"K")
             self.a0 = float(x[0])