private static void DiamondML()
{
var mlContext = new MLContext();
DataRelativePath = $"{BaseDatasetsRelativePath}/diamond_data.csv";
DataPath = GetAbsolutePath(DataRelativePath);
ModelRelativePath = $"{BaseModelsRelativePath}/whitediamondModel.zip";
ModelPath = GetAbsolutePath(ModelRelativePath);
// STEP 1: Common data loading configuration
IDataView dataView = mlContext.Data.LoadFromTextFile <DiamondData>(DataPath, separatorChar: ',', hasHeader: true);
DataOperationsCatalog.TrainTestData trainTestSplit = mlContext.Data.TrainTestSplit(dataView, testFraction: 0.2);
IDataView trainingData = trainTestSplit.TrainSet;
IDataView testData = trainTestSplit.TestSet;
// STEP 2: Common data process configuration with pipeline data transformations
//var dataProcessPipeline = mlContext.Transforms.CopyColumns(outputColumnName: "Features", inputColumnName: nameof(DiamondData.LumiStdHueMean));
var dataProcessPipeline = mlContext.Transforms.Concatenate("Features", "LumiStd", "HueMean");
// STEP 3: Set the training algorithm, then create and config the modelBuilder
var trainer = mlContext.BinaryClassification.Trainers.SdcaLogisticRegression(labelColumnName: "Label", featureColumnName: "Features");
var trainingPipeline = dataProcessPipeline.Append(trainer);
// STEP 4: Train the model fitting to the DataSet
ITransformer trainedModel = trainingPipeline.Fit(trainingData);
// STEP 5: Evaluate the model and show accuracy stats
var predictions = trainedModel.Transform(testData);
var metrics = mlContext.BinaryClassification.Evaluate(data: predictions, labelColumnName: "Label", scoreColumnName: "Score");
ConsoleHelper.PrintBinaryClassificationMetrics(trainer.ToString(), metrics);
// STEP 6: Save/persist the trained model to a .ZIP file
mlContext.Model.Save(trainedModel, trainingData.Schema, ModelPath);
Console.WriteLine("The model is saved to {0}", ModelPath);
// TRY IT: Make a single test prediction, loading the model from .ZIP file
DiamondData sampleDiamond = new DiamondData {
LumiStd = 0.1169f, HueMean = 0.13599f
};
// Create prediction engine related to the loaded trained model
var predEngine = mlContext.Model.CreatePredictionEngine <DiamondData, DiamondPrediction>(trainedModel);
// Score
var resultprediction = predEngine.Predict(sampleDiamond);
Console.WriteLine($"=============== Single Prediction ===============");
Console.WriteLine($"{sampleDiamond.LumiStd}, {sampleDiamond.HueMean} | Prediction: {(Convert.ToBoolean(resultprediction.Prediction) ? "white diamond" : "Non white diamond")} Probability of being white diamond: {resultprediction.Probability} ");
Console.WriteLine($"================End of Process.Hit any key to exit==================================");
}
void FillDiamond(ref Mesh diamond, ref DiamondData data)
{
_vertices[0].pos.x = data.pt1.x;
_vertices[0].pos.y = data.pt1.y;
_vertices[0].uv.x = data.uv1.x;
_vertices[0].uv.y = data.uv1.y;
_vertices[1].pos.x = data.pt2.x;
_vertices[1].pos.y = data.pt2.y;
_vertices[1].uv.x = data.uv2.x;
_vertices[1].uv.y = data.uv2.y;
_vertices[2].pos.x = data.pt3.x;
_vertices[2].pos.y = data.pt3.y;
_vertices[2].uv.x = data.uv3.x;
_vertices[2].uv.y = data.uv3.y;
_vertices[3].pos.x = data.pt4.x;
_vertices[3].pos.y = data.pt4.y;
_vertices[3].uv.x = data.uv4.x;
_vertices[3].uv.y = data.uv4.y;
_vertices[4].pos.x = data.pt5.x;
_vertices[4].pos.y = data.pt5.y;
_vertices[4].uv.x = data.uv5.x;
_vertices[4].uv.y = data.uv5.y;
MeshUpdateFlags flag = GetMeshUpdateFlags();
const int indexCount = 9;
const int vertexCount = 5;
// 設定 Mesh Topologiy
SubMeshDescriptor desc = new SubMeshDescriptor(0, indexCount, MeshTopology.Triangles);
diamond.SetSubMesh(0, desc, flag);
// 宣告 index buffer 結構
diamond.SetIndexBufferParams(indexCount, IndexFormat.UInt16);
// 宣告 vertex buffer 結構
diamond.SetVertexBufferParams(vertexCount, _layouts);
diamond.SetVertexBufferData(_vertices, 0, 0, vertexCount, 0, flag);
diamond.SetIndexBufferData(_indices, 0, 0, indexCount, flag);
}
