cvboost mldata

edit

I trained my data using cvboost algorithm. My input training data contains 10 samples and 100 features. I am writing it into csv file and i am loading from it into mldata object. Call train method using this. But variables from csv file is features+1 all the time which includes response. It trains good but when i call predict(), it throws the following error. PLease someone tell me how to resolve the problem. Here is the error: OpenCV Error: Bad argument (the input sample must be 1d floating-point vector with the same number of elements as the total number of variables or as the number of variables used for training) in predict,