Question

特征列'Features'的架构不匹配：预期的Vector，获得了Vector参数名称：inputSchema 该错误发生在以下代码中

 static readonly string _dataPat=Path.Combine(Environment.CurrentDirectory, "Data", "train_data.csv"); 
static void Main(string[] args)
    {

            var mlContext = new MLContext(seed: 0);

        IDataView data = mlContext.Data.LoadFromTextFile<IrisData>(_dataPath, separatorChar: ',', hasHeader: true);

        string featuresColumnName = "Features";

            var pipeline = mlContext.Transforms
                .Concatenate(featuresColumnName,"Class", "Sex", "AgeGroup", "Embarked")
                .Append(mlContext.Clustering.Trainers.KMeans(featuresColumnName, numberOfClusters: 2));

            var model = pipeline.Fit(data);

错误与代码一致

var model = pipeline.Fit(data);

我的班级IrisData

public class IrisData
{
    [LoadColumn(0)]
    public string Class;

    [LoadColumn(1)]
    public string Sex;

    [LoadColumn(2)]
    public string AgeGroup;

    [LoadColumn(3)]
    public string Embarked;
}

CSV file in my project

Answer 1

您的Features列必须是float s的向量。但是您正在创建一个string s的向量。

您将需要将这些字符串转换为数字。一种方法是使用OneHotEncoding。参见https://docs.microsoft.com/en-us/dotnet/machine-learning/how-to-guides/prepare-data-ml-net#work-with-categorical-data

功能列“功能”的架构不匹配：

1 个答案: