static void Main(String[] args)
{
Console.WriteLine("Hello World!");
var config = new TrainerConfig
{
TrainDataPath = Path.Combine(AppContext.BaseDirectory, "Data", "issues_train.tsv"),
TestDataPath = Path.Combine(AppContext.BaseDirectory, "Data", "issues_test.tsv"),
ModelPath = Path.Combine(AppContext.BaseDirectory, "Models", "github_issues.zip"),
Label = "Area",
}.WithFeature("Title").WithFeature("Description");
var trainer = new Trainer <GitHubIssue, IssuePrediction>(config);
Console.WriteLine("Training...");
var model = trainer.TrainOrLoad();
Console.WriteLine("Training...done!");
Console.WriteLine("Evaluating...");
trainer.Evaluate();
Console.WriteLine("Evaluating...done!");
trainer.Save();
var issue = new GitHubIssue()
{
Title = "WebSockets communication is slow in my machine",
Description = "The WebSockets communication used under the covers by SignalR looks like is going slow in my development machine.."
};
var prediction = trainer.Predict(issue);
Console.WriteLine($"The prediciton is...{prediction.Area}!");
Console.ReadLine();
}
public Trainer(TrainerConfig config)
{
_context = new MLContext(seed: 0);;
_config = config;
}
