public void TestCreateTextInstancesConstant()
{
TestDataset breast = TestDatasets.breastCancerConst;
string trainData = GetDataPath(breast.trainFilename);
var prefix = TestContext.TestName + "-";
string outName = prefix + "BreastCancer.txt";
string statsName = prefix + "BreastCancer.stats.txt";
string outFile = DeleteOutputPath(Dir, outName);
string statsFile = DeleteOutputPath(Dir, statsName);
var argsStr =
string.Format(
"c=CreateInstances {0} {1} cifile={2} cistatsfile={3}",
trainData, breast.extraSettings, outFile, statsFile);
argsStr += " writer=TextInstanceWriter{stats+} disableTracking+";
var args = new TLCArguments();
Assert.IsTrue(CmdParser.ParseArguments(argsStr, args));
RunExperiments.Run(args);
CheckEquality(Dir, outName);
CheckEquality(Dir, statsName);
Done();
}
/// <summary>
/// generic method for parsing arguments using CommandLine. If there's a problem, it throws an InvalidOperationException, with a message giving usage.
/// </summary>
/// <param name="env">The host environment</param>
/// <param name="args">The argument object</param>
/// <param name="settings">The settings string (for example, "threshold-")</param>
/// <param name="name">The name is used for error reporting only</param>
public static void ParseArguments(IHostEnvironment env, object args, string settings, string name = null)
{
if (string.IsNullOrWhiteSpace(settings))
{
return;
}
string errorMsg = null; // non-null errorMsg will indicate an error
try
{
string err = null;
string helpText;
if (!CmdParser.ParseArguments(env, settings, args, e => { err = err ?? e; }, out helpText))
{
errorMsg = err + (!string.IsNullOrWhiteSpace(name) ? "\nUSAGE FOR '" + name + "':\n" : "\nUSAGE:\n") + helpText;
}
}
catch (Exception e)
{
Contracts.Assert(false);
errorMsg = "Unexpected exception thrown while parsing: " + e.Message;
}
if (errorMsg != null)
{
throw Contracts.Except(errorMsg);
}
}
private Instances CreateExtractInstances(string filename, string settings = null)
{
var args = new ExtractInstances.Arguments();
CmdParser.ParseArguments(settings, args);
return(new ExtractInstances(args, filename));
}
protected virtual int HandleListCommand(string[] args, DefaultOptions options)
{
var op = options.List;
ConfigureServices(op);
if (!CmdParser.ParseArguments(args, op))
{
return(1);
}
var command = new NuGetListCommand
{
Url = op.Url,
UserName = op.UserName,
UserPassword = op.UserPassword
};
if (string.IsNullOrEmpty(op.SharePointVersion))
{
command.SharePointVersion = SharePointRuntimVersions.O365;
}
command.Execute();
return(0);
}
/// <summary>
/// Parses arguments using CmdParser. If there's a problem, it throws an InvalidOperationException,
/// with a message giving usage.
/// </summary>
/// <param name="env">The host environment</param>
/// <param name="args">The argument object</param>
/// <param name="settings">The settings string</param>
/// <param name="name">The name is used for error reporting only</param>
private static void ParseArguments(IHostEnvironment env, object args, string settings, string name)
{
Contracts.AssertValue(args);
Contracts.AssertNonEmpty(name);
if (string.IsNullOrWhiteSpace(settings))
{
return;
}
string errorMsg = null;
try
{
string err = null;
string helpText;
if (!CmdParser.ParseArguments(env, settings, args, e => { err = err ?? e; }, out helpText))
{
errorMsg = err + Environment.NewLine + "Usage For '" + name + "':" + Environment.NewLine + helpText;
}
}
catch (Exception e)
{
Contracts.Assert(false);
throw Contracts.Except(e, "Unexpected exception thrown while parsing: " + e.Message);
}
if (errorMsg != null)
{
throw Contracts.Except(errorMsg);
}
}
public void ParseCreateTableArgumentsTestShouldFail(string[] args)
{
var command = CmdParser.ParseArguments(args);
Controller ctrl = new Controller(command);
bool result = ctrl.CheckCommandLineArguments();
Assert.That(result, Is.False);
}
public void CreateTableNonExistingBucketTest()
{
string[] args = new string[] { "create-table", "hopefullynonexistingbucket", testTableName, testFilename, "-t", testToken };
var command = CmdParser.ParseArguments(args);
Controller ctrl = new Controller(command);
var result = ctrl.Execute();
Assert.That(result, Is.False);
}
public void ArgumentParseTest()
{
var env = new ConsoleEnvironment(seed: 42);
var innerArg1 = new SimpleArg()
{
required = -2,
text1 = "}",
text2 = "{",
text3 = " ",
text4 = "\n",
};
var innerArg2 = new SimpleArg()
{
required = -2,
text1 = "{!@# $%^}&*{()}",
text2 = "=",
text3 = "\t",
text4 = @"\\",
};
var innerArg3 = new SimpleArg()
{
required = -2,
text1 = "\r\n",
text2 = "\"",
text3 = "\" \" ",
text4 = "{/a=2 /b=3 /c=4}",
sub1 = CreateComponentFactory("S1", innerArg1.ToString(env)),
sub2 = CreateComponentFactory("S2", innerArg2.ToString(env)),
};
var outerArg1 = new SimpleArg()
{
required = -2,
once = 2,
text2 = "Testing",
text3 = "a=7",
sub1 = CreateComponentFactory("S1", innerArg1.ToString(env)),
sub2 = CreateComponentFactory("S2", innerArg2.ToString(env)),
sub3 = CreateComponentFactory("S3", innerArg3.ToString(env)),
};
var testArg = new SimpleArg();
CmdParser.ParseArguments(env, outerArg1.ToString(env), testArg);
Assert.Equal(outerArg1, testArg);
CmdParser.ParseArguments(env, ((ICommandLineComponentFactory)outerArg1.sub1).GetSettingsString(), testArg = new SimpleArg());
Assert.Equal(innerArg1, testArg);
CmdParser.ParseArguments(env, ((ICommandLineComponentFactory)outerArg1.sub2).GetSettingsString(), testArg = new SimpleArg());
Assert.Equal(innerArg2, testArg);
CmdParser.ParseArguments(env, ((ICommandLineComponentFactory)outerArg1.sub3).GetSettingsString(), testArg = new SimpleArg());
Assert.Equal(innerArg3, testArg);
}
public void ParseCreateTableArgumentsTest(string[] args)
{
var command = CmdParser.ParseArguments(args);
Controller ctrl = new Controller(command);
bool result = ctrl.CheckCommandLineArguments();
Assert.That(result, Is.True);
Assert.That(command, Is.InstanceOf(typeof(CreateTableCommand)));
}
public void BadFileTest()
{
string[] args = new string[] { "create-table", "main", "testtable", "sdfsdfsdf", "-t", testToken };
var command = CmdParser.ParseArguments(args);
Controller ctrl = new Controller(command);
var result = ctrl.Execute();
Assert.That(result, Is.False);
}
public void WriteTableParseArgsOK(Tuple <string[], WriteTableCommand> args)
{
WriteTableCommand command = (WriteTableCommand)CmdParser.ParseArguments(args.Item1);
WriteTableCommand shouldbe = (WriteTableCommand)args.Item2;
// Assert.That(command, Is.InstanceOf(typeof(WriteTableCommand)));
// Assert.AreEqual(command, shouldbe);
command.ShouldHave().AllProperties().EqualTo(shouldbe);
}
public void CreateTableOKTest()
{
string[] args = new string[] { "create-table", testBucketId, testTableName, testFilename, "-t", testToken };
var command = CmdParser.ParseArguments(args);
Controller ctrl = new Controller(command);
var result = ctrl.Execute();
Assert.That(result, Is.True);
}
public void CreateOnExistingTableTestShouldFail()
{
string[] args = new string[] { "create-table", testBucketId, "existingTableTest", testFilename, "-t", testToken };
var command = CmdParser.ParseArguments(args);
Controller ctrl = new Controller(command);
var firstResult = ctrl.Execute();
var secondResult = ctrl.Execute();
Assert.That(secondResult, Is.False);
}
public static IDataView GetKcHouseDataView(string dataPath)
{
var dataSchema = "col=Id:TX:0 col=Date:TX:1 col=Label:R4:2 col=Bedrooms:R4:3 col=Bathrooms:R4:4 col=SqftLiving:R4:5 col=SqftLot:R4:6 col=Floors:R4:7 col=Waterfront:R4:8 col=View:R4:9 col=Condition:R4:10 col=Grade:R4:11 col=SqftAbove:R4:12 col=SqftBasement:R4:13 col=YearBuilt:R4:14 col=YearRenovated:R4:15 col=Zipcode:R4:16 col=Lat:R4:17 col=Long:R4:18 col=SqftLiving15:R4:19 col=SqftLot15:R4:20 header+ sep=,";
var txtArgs = new TextLoader.Arguments();
bool parsed = CmdParser.ParseArguments(s_environment, dataSchema, txtArgs);
s_environment.Assert(parsed);
var txtLoader = new TextLoader(s_environment, txtArgs, new MultiFileSource(dataPath));
return(txtLoader);
}
public void ParseVerboseArgTest(string [] args)
{
//string[] args = new string[] { "-q", "-v" };
///string[] args = new string[] { "create-table", "-f", "asdasd", "-t", "adasd", "-nSomeName", "-b", "sdf", "-v" };
//string [] args = new string[]{ };
var command = CmdParser.ParseArguments(args);
Controller ctrl = new Controller(command);
Assert.That(ctrl._currentLogLevel, Is.EqualTo(LogLevel.debug));
//Assert.That(CmdParser.GlobalOpations.Verbose, Is.True);
}
protected virtual int HandleInstallCommand(string[] args, DefaultOptions options)
{
var op = options.Install;
ConfigureServices(op);
if (!CmdParser.ParseArguments(args, op))
{
return(1);
}
var sources = new List <string>();
if (!string.IsNullOrEmpty(op.Source))
{
sources.Add(op.Source);
}
sources.AddRange(DefaultValues.DefaultNuGetRepositories);
var command = new DefaultInstallCommand
{
PackageSources = sources,
Url = op.Url,
Id = op.Id,
Version = op.Version,
PreRelease = op.PreRelease,
UserName = op.UserName,
UserPassword = op.UserPassword,
SharePointVersion = op.SharePointVersion,
SharePointEdition = op.SharePointEdition,
SharePointApi = op.SharePointApi,
ToolId = op.ToolId,
ToolVersion = op.ToolVersion,
Force = op.Force
};
if (string.IsNullOrEmpty(op.SharePointVersion))
{
command.SharePointVersion = SharePointRuntimVersions.O365;
}
command.Execute();
return(0);
}
public void TestCreateTextInstances()
{
TestDataset adult = TestDatasets.adult;
string trainData = GetDataPath(adult.trainFilename);
string testData = GetDataPath(adult.testFilename);
var prefix = TestContext.TestName + "-";
string outName = prefix + "Adult-Train.txt";
string statsName = prefix + "Adult-Train.stats.txt";
string outTestName = prefix + "Adult-Test.txt";
string testStatsName = prefix + "Adult-Test.stats.txt";
string outValidName = prefix + "Adult-Valid.txt";
string validStatsName = prefix + "Adult-Valid.stats.txt";
string outFile = DeleteOutputPath(Dir, outName);
string statsFile = DeleteOutputPath(Dir, statsName);
string outTestFile = DeleteOutputPath(Dir, outTestName);
string testStatsFile = DeleteOutputPath(Dir, testStatsName);
string outValidFile = DeleteOutputPath(Dir, outValidName);
string validStatsFile = DeleteOutputPath(Dir, validStatsName);
var argsStr =
string.Format(
"/c=CreateInstances {0} /test={1} /valid={1} /cacheinst=- {2} " +
"/cifile={3} /cistatsfile={4} /citestfile={5} /citeststatsfile={6} /civalidfile={7} /civalidstatsfile={8}",
trainData, testData, adult.extraSettings,
outFile, statsFile, outTestFile, testStatsFile, outValidFile, validStatsFile);
argsStr += " /writer TextInstanceWriter{/stats=+} /disableTracking=+";
var args = new TLCArguments();
Assert.IsTrue(CmdParser.ParseArguments(argsStr, args));
RunExperiments.Run(args);
CheckEquality(Dir, outName);
CheckEquality(Dir, statsName);
CheckEquality(Dir, outTestName);
CheckEquality(Dir, testStatsName);
CheckEquality(Dir, outValidName);
CheckEquality(Dir, validStatsName);
Done();
}
public void TestPcaTransform()
{
// Force Microsoft.ML.Runtime.PCA assembly to be loaded into the AppDomain so
// ReflectionUtils.FindClassCore does not return null when called by ReflectionUtils.CreateInstance
Assert.AreEqual(typeof(PCAPredictor).Name, "PCAPredictor");
string trainData = GetDataPath(TestDatasets.mnistTiny28.trainFilename);
string fileName = TestContext.TestName + "-Train.txt";
string outFile = DeleteOutputPath(Dir, fileName);
const int rank = 3;
string pcaTransformArgs = string.Format("/inst Trans{{trans=pca {{k={0} seed=1}}}}", rank);
var argsStr1 = string.Format(
"/c CreateInstances {0} {1} /rs=1 /cifile={2}",
trainData,
pcaTransformArgs,
outFile);
var args1 = new TLCArguments();
Assert.IsTrue(CmdParser.ParseArguments(argsStr1, args1));
RunExperiments.Run(args1);
CheckEquality(Dir, fileName);
// Verify the scales of the transformed features decrease with respect to the feature index
TlcTextInstances outputInstances = new TlcTextInstances(new TlcTextInstances.Arguments(), outFile);
Double[] l1norms = new Double[rank];
foreach (Instance instance in outputInstances)
{
Assert.IsTrue(instance.Features.Count == rank);
for (int i = 0; i < instance.Features.Values.Length; i++)
{
l1norms[i] += (instance.Features.Values[i] < 0 ? -instance.Features.Values[i] : instance.Features.Values[i]);
}
}
for (int i = 0; i < l1norms.Length - 1; i++)
{
Assert.IsTrue(l1norms[i] > l1norms[i + 1]);
}
Done();
}
public void TestCreateTextInstancesWithNormalization()
{
TestDataset dataset = TestDatasets.mnistTiny28;
string trainData = GetDataPath(dataset.trainFilename);
string testData = GetDataPath(dataset.testFilename);
var prefix = TestContext.TestName + "-";
string outFile1 = DeleteOutputPath(Dir, prefix + "Norm-Separate-Train.txt");
string outTestFile1 = DeleteOutputPath(Dir, prefix + "Norm-Separate-Test.txt");
string outFile2 = DeleteOutputPath(Dir, prefix + "Norm-Trans-Train.txt");
string outTestFile2 = DeleteOutputPath(Dir, prefix + "Norm-Trans-Test.txt");
string transArgs = "inst=Trans{trans=RFF {rng=1}}";
var argsStr1 = string.Format(
"/c=CreateInstances {0} /test={1} /norm=MinMaxNormalizer /{2} /cifile={3} /citestfile={4}",
trainData, testData, transArgs, outFile1, outTestFile1);
var args1 = new TLCArguments();
Assert.IsTrue(CmdParser.ParseArguments(argsStr1, args1));
RunExperiments.Run(args1);
var argsStr2 = string.Format(
"/c=CreateInstances {0} /test={1} /inst Trans{{trans=MinMaxNormalizer {2}}} /cifile={3} /citestfile={4}",
trainData, testData, transArgs, outFile2, outTestFile2);
var args2 = new TLCArguments();
Assert.IsTrue(CmdParser.ParseArguments(argsStr2, args2));
RunExperiments.Run(args2);
var instances1 = new TlcTextInstances(new TlcTextInstances.Arguments(), outFile1);
var instances2 = new TlcTextInstances(new TlcTextInstances.Arguments(), outFile2);
CompareInstances(instances1, instances2);
var testInstances1 = new TlcTextInstances(new TlcTextInstances.Arguments(), outTestFile1);
var testInstances2 = new TlcTextInstances(new TlcTextInstances.Arguments(), outTestFile2);
CompareInstances(testInstances1, testInstances2);
Done();
}
// [TestCase("blablabla")]
public void UpdateNonExistingTableTest(string nonExistingTableId)
{
string updateFileName = "filetoupdate" + DateTime.Now.Ticks + ".csv";
string[] args = new string[] { "write-table", nonExistingTableId, updateFileName, "-t", testToken };
//creates test file
FileInfo updateFileInfo = new FileInfo(updateFileName);
string[] dataToUpdate = PrepareDataToUpdate(100);
System.IO.File.WriteAllLines(updateFileInfo.FullName, dataToUpdate);
var command = CmdParser.ParseArguments(args);
Controller ctrl = new Controller(command);
bool result = ctrl.Execute();
Assert.That(result, Is.False);
updateFileInfo.Delete();
}
/// <summary>
/// Process a script to be parsed (from the input resource).
/// </summary>
private static void Process(IndentingTextWriter wrt, string text, ArgsBase defaults)
{
var env = new TlcEnvironment(seed: 42);
using (wrt.Nest())
{
var args1 = defaults.Clone();
using (wrt.Nest())
{
if (!CmdParser.ParseArguments(env, text, args1, s => wrt.WriteLine("*** {0}", s)))
{
wrt.WriteLine("*** Failed!");
}
}
string str1 = args1.ToString();
wrt.WriteLine("ToString: {0}", str1);
string settings1 = CmdParser.GetSettings(env, args1, defaults, SettingsFlags.None);
wrt.WriteLine("Settings: {0}", settings1);
var args2 = defaults.Clone();
using (wrt.Nest())
{
if (!CmdParser.ParseArguments(env, settings1, args2, s => wrt.WriteLine("*** BUG: {0}", s)))
{
wrt.WriteLine("*** BUG: parsing result of GetSettings failed!");
}
}
string str2 = args2.ToString();
if (str1 != str2)
{
wrt.WriteLine("*** BUG: ToString Mismatch: {0}", str2);
}
string settings2 = CmdParser.GetSettings(env, args2, defaults, SettingsFlags.None);
if (settings1 != settings2)
{
wrt.WriteLine("*** BUG: Settings Mismatch: {0}", settings2);
}
}
}
public void TestCrossValidationWithInvalidTester()
{
var argsStr = GetDataPath(TestDatasets.breastCancer.trainFilename)
+ " /ev=MulticlassTester /o z.txt /threads=+ /disableTracking=+";
var args = new TLCArguments();
CmdParser.ParseArguments(argsStr, args);
try
{
RunExperiments.Run(args);
}
catch (AggregateException ex)
{
Log("Caught expected exception: {0}", ex);
Done();
return;
}
Fail("Expected exception!");
Done();
}
protected virtual int HandleUpdateCommand(string[] args, DefaultOptions options)
{
var op = options.Update;
ConfigureServices(op);
if (!CmdParser.ParseArguments(args, op))
{
return(1);
}
var command = new DefaultUpdateCommand
{
Source = op.Source,
Url = op.Url,
Id = op.Id,
Version = op.Version,
PreRelease = op.PreRelease,
UserName = op.UserName,
UserPassword = op.UserPassword,
SharePointVersion = op.SharePointVersion,
SharePointEdition = op.SharePointEdition,
SharePointApi = op.SharePointApi,
Force = op.Force
};
if (string.IsNullOrEmpty(op.SharePointVersion))
{
command.SharePointVersion = SharePointRuntimVersions.O365;
}
command.Execute();
return(0);
}
protected virtual int HandlePushCommand(string[] args, DefaultOptions options)
{
var op = options.Push;
ConfigureServices(op);
if (!CmdParser.ParseArguments(args, op))
{
return(1);
}
Info(string.Format("Resolving package path [{0}]", op.Package));
var packageFileFullPath = Path.GetFullPath(op.Package);
Info(string.Format("Resolved package path [{0}] into [{1}]", op.Package, packageFileFullPath));
if (!File.Exists(packageFileFullPath))
{
Info(string.Format("File does not exist:[{0}]", packageFileFullPath));
}
using (var stream = File.Open(packageFileFullPath, FileMode.Open))
{
var command = new DefaultNuGetPushCommand
{
Source = op.Source,
ApiKey = op.ApiKey,
Package = stream
};
command.Execute();
}
return(0);
}
public void TestFeatureHandlerIncorrectMapping()
{
string trainData = GetDataPath(TestDatasets.breastCancer.trainFilename);
string dataModelFile = DeleteOutputPath(Dir, TestContext.TestName + "-data-model.zip");
string ciFile = DeleteOutputPath(Dir, TestContext.TestName + "-ci.tsv");
string argsString = string.Format(
"/c CreateInstances {0} /inst Text{{text=1,2,3}} /m {1} /cifile {2}",
trainData,
dataModelFile,
ciFile);
var args = new TLCArguments();
Assert.IsTrue(CmdParser.ParseArguments(argsString, args));
RunExperiments.Run(args);
string ciFailFile = DeleteOutputPath(Dir, TestContext.TestName + "-ci-fail.tsv");
argsString = string.Format(
"/c CreateInstances {0} /inst Text{{text=1,2}} /im {1} /cifile {2}",
trainData,
dataModelFile,
ciFailFile);
args = new TLCArguments();
Assert.IsTrue(CmdParser.ParseArguments(argsString, args));
try
{
RunExperiments.Run(args);
Assert.Fail("Expected to throw with different input model format");
}
catch (Exception ex)
{
Assert.IsTrue(ex.GetBaseException() is InvalidOperationException);
}
Done();
}
public int Run(string[] args)
{
Args = args;
if (!args.Any())
{
args = new string[1] {
"help"
}
}
;
var result = 0;
var options = new DefaultOptions();
// get the log file location
var preDefaultArgs = new EmptySubOptionsBase();
CmdParser.ParseArguments(args, preDefaultArgs);
if (!string.IsNullOrEmpty(preDefaultArgs.LogFile))
{
LogFile = preDefaultArgs.LogFile;
}
if (!string.IsNullOrEmpty(preDefaultArgs.Output) && preDefaultArgs.Output.ToLower() == "json")
{
IsJsonOutput = true;
}
if (!CmdParser.ParseArguments(args, options, (verb, subOption) =>
{
if (!hadFirstRun)
{
ConfigureServiceContainer();
AppDomain.CurrentDomain.AssemblyResolve += CurrentDomain_AssemblyResolve;
hadFirstRun = true;
}
var currentFilePath = GetType().Assembly.Location;
var currentFolderPath = new DirectoryInfo(currentFilePath).Parent.FullName;
Info(string.Format("Metapack client v{0}", GetCurrentClientVersion()));
Info(string.Format("Working directory: [{0}]", currentFolderPath));
WorkingDirectory = currentFolderPath;
ParseAppConfig();
if (options.List != null)
{
result = HandleListCommand(args, options);
}
else if (options.Install != null)
{
result = HandleInstallCommand(args, options);
}
else if (options.Update != null)
{
result = HandleUpdateCommand(args, options);
}
else if (options.Push != null)
{
result = HandlePushCommand(args, options);
}
else if (options.Version != null)
{
result = HandleVersionCommand(args, options);
}
else if (IsHelp)
{
Info(options.GetUsage());
result = 0;
}
else
{
result = HandleMissedCommand(options);
}
}))
{
if (!IsHelp)
{
HandleWrongArgumentParsing();
}
}
return(result);
}
public void FactoryExampleTest()
{
var dataset = TestDatasets.adultText;
string dataFilename = GetDataPath(dataset.trainFilename);
string testDataFilename = GetDataPath(dataset.testFilename);
///********* Training a model *******//
string modelFilename = Path.GetTempFileName();
TLCArguments cmd = new TLCArguments();
Assert.True(CmdParser.ParseArguments(dataset.extraSettings, cmd));
cmd.command = Command.Train;
cmd.modelfile = modelFilename;
cmd.datafile = dataFilename;
cmd.instancesSettings = dataset.settings;
cmd.classifierName = TestLearners.linearSVM.Trainer;
RunExperiments.Run(cmd);
// Load and make predictions with a previously saved model.
IDataModel dataModel;
IDataStats dataStats;
var predictor = (IDistributionPredictor <Instance, Float, Float>)PredictorUtils.LoadPredictor(
out dataModel, out dataStats, modelFilename);
var instanceFactory = ReflectionUtilsOld.CreateInstanceOld <IInstanceFactory, SignatureInstances>(
cmd.instancesClass, cmd.instancesSettings, null, dataModel);
bool headerSkip = true;
List <Float> outputs = new List <Float>();
List <Float> probabilities = new List <Float>();
using (StreamReader reader = new StreamReader(testDataFilename))
{
List <string> features = new List <string>();
string text;
long line = 0;
while ((text = reader.ReadLine()) != null)
{
++line;
if (string.IsNullOrWhiteSpace(text))
{
continue;
}
string[] cols = text.Split(',');
Assert.True(cols.Length == 15);
if (headerSkip)
{
// skip header line
headerSkip = false;
continue;
}
features.Clear();
// Add in the "max dimensionality"
features.Add("15");
for (int col = 0; col < cols.Length; col++)
{
string s = cols[col].Trim();
switch (col)
{
case 0:
case 2:
case 4:
case 10:
case 11:
case 12:
case 14:
// numeric feature or label -- add if non-zero
Float val = InstancesUtils.FloatParse(s);
if (val == 0) // Beware of NaNs - they should be recorded!
{
continue;
}
break;
}
features.Add(col + ":" + s);
}
Instance instance = instanceFactory.ProduceInstance(line, features.ToArray());
Float rawOutput, probability;
probability = predictor.PredictDistribution(instance, out rawOutput);
outputs.Add(rawOutput);
probabilities.Add(probability);
}
}
List <Float> originalOutputs = new List <Float>();
List <Float> originalProbabilities = new List <Float>();
var env = new LocalEnvironment(SysRandom.Wrap(RunExperiments.GetRandom(cmd)));
Instances instances = RunExperiments.CreateTestData(cmd, testDataFilename, dataModel, null, env);
foreach (Instance instance in instances)
{
Float rawOutput, probability;
probability = predictor.PredictDistribution(instance, out rawOutput);
originalOutputs.Add(rawOutput);
originalProbabilities.Add(probability);
}
CollectionAssert.Equal(outputs, originalOutputs);
CollectionAssert.Equal(probabilities, originalProbabilities);
File.Delete(modelFilename);
Done();
}
// This method is called if only a datafile is specified, without a loader/term and value columns.
// It determines the type of the Value column and returns the appropriate TextLoader component factory.
private static IComponentFactory <IMultiStreamSource, IDataLoader> GetLoaderFactory(string filename, bool keyValues, IHost host)
{
Contracts.AssertValue(host);
// If the user specified non-key values, we define the value column to be numeric.
if (!keyValues)
{
return(ComponentFactoryUtils.CreateFromFunction <IMultiStreamSource, IDataLoader>(
(env, files) => TextLoader.Create(
env,
new TextLoader.Arguments()
{
Column = new[]
{
new TextLoader.Column("Term", DataKind.TX, 0),
new TextLoader.Column("Value", DataKind.Num, 1)
}
},
files)));
}
// If the user specified key values, we scan the values to determine the range of the key type.
ulong min = ulong.MaxValue;
ulong max = ulong.MinValue;
try
{
var txtArgs = new TextLoader.Arguments();
bool parsed = CmdParser.ParseArguments(host, "col=Term:TX:0 col=Value:TX:1", txtArgs);
host.Assert(parsed);
var data = TextLoader.ReadFile(host, txtArgs, new MultiFileSource(filename));
using (var cursor = data.GetRowCursor(c => true))
{
var getTerm = cursor.GetGetter <DvText>(0);
var getVal = cursor.GetGetter <DvText>(1);
DvText txt = default(DvText);
using (var ch = host.Start("Creating Text Lookup Loader"))
{
long countNonKeys = 0;
while (cursor.MoveNext())
{
getVal(ref txt);
ulong res;
// Try to parse the text as a key value between 1 and ulong.MaxValue. If this succeeds and res>0,
// we update max and min accordingly. If res==0 it means the value is missing, in which case we ignore it for
// computing max and min.
if (Conversions.Instance.TryParseKey(ref txt, 1, ulong.MaxValue, out res))
{
if (res < min && res != 0)
{
min = res;
}
if (res > max)
{
max = res;
}
}
// If parsing as key did not succeed, the value can still be 0, so we try parsing it as a ulong. If it succeeds,
// then the value is 0, and we update min accordingly.
else if (Conversions.Instance.TryParse(ref txt, out res))
{
ch.Assert(res == 0);
min = 0;
}
//If parsing as a ulong fails, we increment the counter for the non-key values.
else
{
var term = default(DvText);
getTerm(ref term);
if (countNonKeys < 5)
{
ch.Warning("Term '{0}' in mapping file is mapped to non key value '{1}'", term, txt);
}
countNonKeys++;
}
}
if (countNonKeys > 0)
{
ch.Warning("Found {0} non key values in the file '{1}'", countNonKeys, filename);
}
if (min > max)
{
min = 0;
max = uint.MaxValue - 1;
ch.Warning("did not find any valid key values in the file '{0}'", filename);
}
else
{
ch.Info("Found key values in the range {0} to {1} in the file '{2}'", min, max, filename);
}
ch.Done();
}
}
}
catch (Exception e)
{
throw host.Except(e, "Failed to parse the lookup file '{0}' in TermLookupTransform", filename);
}
TextLoader.Column valueColumn = new TextLoader.Column("Value", DataKind.U4, 1);
if (max - min < (ulong)int.MaxValue)
{
valueColumn.KeyRange = new KeyRange(min, max);
}
else if (max - min < (ulong)uint.MaxValue)
{
valueColumn.KeyRange = new KeyRange(min);
}
else
{
valueColumn.Type = DataKind.U8;
valueColumn.KeyRange = new KeyRange(min);
}
return(ComponentFactoryUtils.CreateFromFunction <IMultiStreamSource, IDataLoader>(
(env, files) => TextLoader.Create(
env,
new TextLoader.Arguments()
{
Column = new[]
{
new TextLoader.Column("Term", DataKind.TX, 0),
valueColumn
}
},
files)));
}
private static void RunGraphCore(EnvironmentBlock *penv, IHostEnvironment env, string graphStr, int cdata, DataSourceBlock **ppdata)
{
Contracts.AssertValue(env);
var args = new RunGraphArgs();
string err = null;
if (!CmdParser.ParseArguments(env, graphStr, args, e => err = err ?? e))
{
throw env.Except(err);
}
int?maxThreadsAllowed = Math.Min(args.parallel > 0 ? args.parallel.Value : penv->maxThreadsAllowed, penv->maxThreadsAllowed);
maxThreadsAllowed = penv->maxThreadsAllowed > 0 ? maxThreadsAllowed : args.parallel;
var host = env.Register("RunGraph", args.randomSeed, null);
JObject graph;
try
{
graph = JObject.Parse(args.graph);
}
catch (JsonReaderException ex)
{
throw host.Except(ex, "Failed to parse experiment graph: {0}", ex.Message);
}
var runner = new GraphRunner(host, graph["nodes"] as JArray);
var dvNative = new IDataView[cdata];
try
{
for (int i = 0; i < cdata; i++)
{
dvNative[i] = new NativeDataView(host, ppdata[i]);
}
// Setting inputs.
var jInputs = graph["inputs"] as JObject;
if (graph["inputs"] != null && jInputs == null)
{
throw host.Except("Unexpected value for 'inputs': {0}", graph["inputs"]);
}
int iDv = 0;
if (jInputs != null)
{
foreach (var kvp in jInputs)
{
var pathValue = kvp.Value as JValue;
if (pathValue == null)
{
throw host.Except("Invalid value for input: {0}", kvp.Value);
}
var path = pathValue.Value <string>();
var varName = kvp.Key;
var type = runner.GetPortDataKind(varName);
switch (type)
{
case TlcModule.DataKind.FileHandle:
var fh = new SimpleFileHandle(host, path, false, false);
runner.SetInput(varName, fh);
break;
case TlcModule.DataKind.DataView:
IDataView dv;
if (!string.IsNullOrWhiteSpace(path))
{
var extension = Path.GetExtension(path);
if (extension == ".txt")
{
dv = TextLoader.LoadFile(host, new TextLoader.Options(), new MultiFileSource(path));
}
else if (extension == ".dprep")
{
dv = DataFlow.FromDPrepFile(path).ToDataView();
}
else
{
dv = new BinaryLoader(host, new BinaryLoader.Arguments(), path);
}
}
else
{
Contracts.Assert(iDv < dvNative.Length);
// prefetch all columns
dv = dvNative[iDv++];
var prefetch = new int[dv.Schema.Count];
for (int i = 0; i < prefetch.Length; i++)
{
prefetch[i] = i;
}
dv = new CacheDataView(host, dv, prefetch);
}
runner.SetInput(varName, dv);
break;
case TlcModule.DataKind.PredictorModel:
PredictorModel pm;
if (!string.IsNullOrWhiteSpace(path))
{
using (var fs = File.OpenRead(path))
pm = new PredictorModelImpl(host, fs);
}
else
{
throw host.Except("Model must be loaded from a file");
}
runner.SetInput(varName, pm);
break;
case TlcModule.DataKind.TransformModel:
TransformModel tm;
if (!string.IsNullOrWhiteSpace(path))
{
using (var fs = File.OpenRead(path))
tm = new TransformModelImpl(host, fs);
}
else
{
throw host.Except("Model must be loaded from a file");
}
runner.SetInput(varName, tm);
break;
default:
throw host.Except("Port type {0} not supported", type);
}
}
}
runner.RunAll();
// Reading outputs.
using (var ch = host.Start("Reading outputs"))
{
var jOutputs = graph["outputs"] as JObject;
if (jOutputs != null)
{
foreach (var kvp in jOutputs)
{
var pathValue = kvp.Value as JValue;
if (pathValue == null)
{
throw host.Except("Invalid value for input: {0}", kvp.Value);
}
var path = pathValue.Value <string>();
var varName = kvp.Key;
var type = runner.GetPortDataKind(varName);
switch (type)
{
case TlcModule.DataKind.FileHandle:
var fh = runner.GetOutput <IFileHandle>(varName);
throw host.ExceptNotSupp("File handle outputs not yet supported.");
case TlcModule.DataKind.DataView:
var idv = runner.GetOutput <IDataView>(varName);
if (!string.IsNullOrWhiteSpace(path))
{
SaveIdvToFile(idv, path, host);
}
else
{
var infos = ProcessColumns(ref idv, args.maxSlots, host);
SendViewToNative(ch, penv, idv, infos);
}
break;
case TlcModule.DataKind.PredictorModel:
var pm = runner.GetOutput <PredictorModel>(varName);
if (!string.IsNullOrWhiteSpace(path))
{
SavePredictorModelToFile(pm, path, host);
}
else
{
throw host.Except("Returning in-memory models is not supported");
}
break;
case TlcModule.DataKind.TransformModel:
var tm = runner.GetOutput <TransformModel>(varName);
if (!string.IsNullOrWhiteSpace(path))
{
using (var fs = File.OpenWrite(path))
tm.Save(host, fs);
}
else
{
throw host.Except("Returning in-memory models is not supported");
}
break;
case TlcModule.DataKind.Array:
var objArray = runner.GetOutput <object[]>(varName);
if (objArray is PredictorModel[])
{
var modelArray = (PredictorModel[])objArray;
// Save each model separately
for (var i = 0; i < modelArray.Length; i++)
{
var modelPath = string.Format(CultureInfo.InvariantCulture, path, i);
SavePredictorModelToFile(modelArray[i], modelPath, host);
}
}
else
{
throw host.Except("DataKind.Array type {0} not supported", objArray.First().GetType());
}
break;
default:
throw host.Except("Port type {0} not supported", type);
}
}
}
}
}
finally
{
// The raw data view is disposable so it lets go of unmanaged raw pointers before we return.
for (int i = 0; i < dvNative.Length; i++)
{
var view = dvNative[i];
if (view == null)
{
continue;
}
host.Assert(view is IDisposable);
var disp = (IDisposable)dvNative[i];
disp.Dispose();
}
}
}
// This method is called if only a datafile is specified, without a loader/term and value columns.
// It determines the type of the Value column and returns the appropriate TextLoader component factory.
private static IComponentFactory <IMultiStreamSource, IDataLoader> GetLoaderFactory(string filename, bool keyValues, IHost host)
{
Contracts.AssertValue(host);
// If the user specified non-key values, we define the value column to be numeric.
if (!keyValues)
{
return(ComponentFactoryUtils.CreateFromFunction <IMultiStreamSource, IDataLoader>(
(env, files) => TextLoader.Create(
env,
new TextLoader.Arguments()
{
Column = new[]
{
new TextLoader.Column("Term", DataKind.TX, 0),
new TextLoader.Column("Value", DataKind.Num, 1)
}
},
files)));
}
// If the user specified key values, we scan the values to determine the range of the key type.
ulong min = ulong.MaxValue;
ulong max = ulong.MinValue;
try
{
var txtArgs = new TextLoader.Arguments();
bool parsed = CmdParser.ParseArguments(host, "col=Term:TX:0 col=Value:TX:1", txtArgs);
host.Assert(parsed);
var data = TextLoader.ReadFile(host, txtArgs, new MultiFileSource(filename));
using (var cursor = data.GetRowCursor(c => true))
{
var getTerm = cursor.GetGetter <ReadOnlyMemory <char> >(0);
var getVal = cursor.GetGetter <ReadOnlyMemory <char> >(1);
ReadOnlyMemory <char> txt = default;
using (var ch = host.Start("Creating Text Lookup Loader"))
{
long countNonKeys = 0;
while (cursor.MoveNext())
{
getVal(ref txt);
ulong res;
// Try to parse the text as a key value between 1 and ulong.MaxValue. If this succeeds and res>0,
// we update max and min accordingly. If res==0 it means the value is missing, in which case we ignore it for
// computing max and min.
if (Runtime.Data.Conversion.Conversions.Instance.TryParseKey(in txt, 1, ulong.MaxValue, out res))
{
if (res < min && res != 0)
{
min = res;
}
if (res > max)
{
max = res;
}
}
// If parsing as key did not succeed, the value can still be 0, so we try parsing it as a ulong. If it succeeds,
// then the value is 0, and we update min accordingly.
else if (Runtime.Data.Conversion.Conversions.Instance.TryParse(in txt, out res))
{
ch.Assert(res == 0);
min = 0;
}