When writing teacher Wu Enda’s machine learning class, there was a problem with the result of using scipy.optimize.fmin_cg

Tips for training Completion:

Warning: Desired error not necessarily achieved due to precision loss.
         Current function value:

The reason is I in front of def computeCost (myTheta myX, myy, mylambda = 0.) : # cost function first parameter myTheta write wrong, right should be myTheta… This problem has been looking for an hour, sad, it is good to change

Read More: