private void btn_CheckResult_Click(object sender, EventArgs e)
{
long len = long.Parse(this.txt_DataLength.Text);
int deep = int.Parse(this.txt_LearnDeep.Text);
ExpectList <T> el = new ExpectReader().ReadHistory <T>(long.Parse(this.txt_BegExpect.Text), len + deep + 1);
for (int i = 0; i < 10; i++)
{
MachineLearnClass <int, int> SelectFunc = null;
MLType = (Type)this.ddl_MLFunc.SelectedValue;
SelectFunc = (MachineLearnClass <int, int>)ClassOperateTool.getInstanceByType(MLType);
//暂时屏蔽机器学习功能
//TestSet = new MLDataFactory(el).getAllSpecColRoundLabelAndFeatures(i,deep, chkb_AllUseShift.Checked ? 1 : 0);
SelectFunc.OnLoadLocalFile = GetLocalFile;
SelectFunc.LoadSummary();
SelectFunc.FillStructBySummary(i);
SelectFunc.SetTestInstances(TestSet);
this.Cursor = Cursors.WaitCursor;
ExecClass ec = new ExecClass();
ec.GroupId = (i + 1) % 10;
ec.MaxEnt = SelectFunc;
ec.TestData = TestSet;
new Thread(new ThreadStart(ec.Exec)).Start();
}
}
private void btn_Train_Click(object sender, EventArgs e)
{
//return;//暂时不支持训练集回测
long len = long.Parse(this.txt_DataLength.Text);
int deep = int.Parse(this.txt_LearnDeep.Text);
ExpectList <TimeSerialData> el = new ExpectReader().ReadHistory <TimeSerialData>(long.Parse(this.txt_BegExpect.Text), len + deep + 1);
//MLDataFactory mldf = new MLDataFactory(ExpectList.getExpectList(el));
DataCategroyType = (Type)this.ddl_categryFunc.SelectedValue;
MLType = (Type)this.ddl_MLFunc.SelectedValue;
MLDataCategoryFactoryClass mldf = (MLDataCategoryFactoryClass)ClassOperateTool.getInstanceByType(DataCategroyType);
mldf.Init(ExpectList.getExpectList(el));
for (int i = 0; i < 10; i++)
{
//MLInstances<int, int> TrainSet = mldf.getAllSpecColRoundLabelAndFeatures(i,deep, chkb_AllUseShift.Checked ? 1 : 0);
MLInstances <int, int> TrainSet = mldf.getCategoryData(i, deep, chkb_AllUseShift.Checked ? 1 : 0);
MachineLearnClass <int, int> SelectFunc = (MachineLearnClass <int, int>)ClassOperateTool.getInstanceByType(MLType);//获取机器学习类型
SelectFunc.OnTrainFinished += OnTrainFinished;
SelectFunc.OnPeriodEvent += OnPeriodEvent;
SelectFunc.OnSaveEvent += SaveData;
SelectFunc.GroupId = i;
SelectFunc.FillTrainData(TrainSet);
SelectFunc.InitTrain();
SelectFunc.TrainIterorCnt = int.Parse(txt_IteratCnt.Text);
SelectFuncs.Add(SelectFunc);
this.txt_begT.Text = DateTime.Now.ToLongTimeString();
this.Cursor = Cursors.WaitCursor;
RunningThread = new Thread(SelectFunc.Train);
RunningThread.Start();
ThreadCnt++;
}
}
public frm_TrainForm()
{
InitializeComponent();
DataTable dt = ClassOperateTool.getAllSubClass(typeof(MachineLearnClass <int, int>), "text", "value");
this.ddl_MLFunc.DataSource = dt;
this.ddl_MLFunc.DisplayMember = "text";
this.ddl_MLFunc.ValueMember = "value";
DataTable dt_categery = ClassOperateTool.getAllSubClass(typeof(MLDataFactory), "", "");
}