zgyao / decisiontree_c4.5 Goto Github PK

Hi zgyao,

I have implemented your code and got it executing, but with the golf example it only shows the outcome as "Play", "Don't Play". How can I see the associated confidence factor for each outcome ?

Thanks !

David (dnettlet)

When processing the attribute information entropy of continuous value data ( the ComputeAttributeEntropy() function), the code directly calculates the original value of the original continuous value. In fact, should we first divide two subsets of data that are greater than and less than the partition point based on the partition point, and then calculate the GainRatio by counting the number of data in the two subsets according to the two terms
当处理连续值数据的属性信息熵（即ComputeAttributeEntropy()函数），代码中是直接按照原先连续值的原始取值计算的。事实上是不是应该先根据划分点划分出大于和小于划分点的两个数据子集后，再统计两个子集的数据个数按照两项来算GainRatio

Hi,

I am trying to compile "DecisionTree_C4.5" source code in Visual Studio 13 by creating an empty project. I find the following errors and warning messages. Can you please help me to solve this problem.
Thank you.

Errors:
Error 12 error C2057: expected constant expression
int Count[Unique_Values.size()] = {0};

Error 13 error C2466: cannot allocate an array of constant size 0
int Count[Unique_Values.size()] = {0};

Error 17 error C2057: expected constant expression
int Count[Score_Range.size()] = {0};

Error 18 error C2466: cannot allocate an array of constant size 0
int Count[Score_Range.size()] = {0};

Warning:
Warning 30 warning C4018: '<' : signed/unsigned mismatch
Warning 7 warning C4554: '&' : check operator precedence for possible error; use parentheses to clarify precedence