__init__.py

import _adolc
from _adolc import *
from wrapped_functions import * 
import cgraph

try:
    import sparse
except:
    print 'adolc Notice: sparse drivers not available'
    
try:
    import colpack

except:
    print 'adolc Notice: colpack drivers not available'

# testing
from numpy.testing import Tester
test = Tester().test

__doc__ = """
Adolc: Algorithmic Differentiation Software 
    see http://www.math.tu-dresden.de/~adol-c/ for documentation of Adolc 
    http://github.com/b45ch1/pyadolc/tree/master for more information and documentation of this Python extension
    
    return values are always numpy arrays!
    
    Example Session: 
    from numpy import *
    from adolc import * 
    def vector_f(x): 
    \tV=vander(x) 
    \treturn dot(v,x)

    x = arange(5,dtype=float)
    ax = adouble(x)
    
    trace_on(0)
    independent(ax)
    ay = vector_f(ax)
    dependent(ay)
    trace_off()

    x = array([1.,4.,0.,0.,0.])
    y = function(0,x)
    J = jacobian(0,x)
"""