Was trying to fit isotonic regression with non-trivial y_min and y_max:

In [17]: X

Out[17]: 

array([ 1.26336413,  1.31853693, -0.57200917,  0.3072928 , -0.70686507,

       -0.17614937, -1.59943059,  1.05908504,  1.3958263 ,  1.90580318,

        0.20992272,  0.02836316, -0.08092235,  0.44438247,  0.01791253,

       -0.3771914 , -0.89577538, -0.37726249, -1.32687569,  0.18013201])

In [18]: iso.isotonic_regression(X, y_min=0, y_max=0.1)

Out[18]: 

array([-0.00826919, -0.00826919, -0.00826919, -0.00826919, -0.00826919,

       -0.00826919, -0.00826919,  0.10449344,  0.10449344,  0.10449344,

        0.10449344,  0.10449344,  0.10449344,  0.10449344,  0.10449344,

        0.10449344,  0.10449344,  0.10449344,  0.10449344,  0.10449344])

The solution does not satisfy the bounds that each entry should be in [0,0.1]

--

Jonathan Taylor                         
Dept. of Statistics                     
Sequoia Hall, 137                         
390 Serra Mall
Stanford, CA 94305
Tel:   650.723.9230
Fax:   650.725.8977
Web: http://www-stat.stanford.edu/~jtaylo