internal BatchPredictionEngine(IHostEnvironment env, Stream modelStream, bool ignoreMissingColumns,
SchemaDefinition inputSchemaDefinition = null, SchemaDefinition outputSchemaDefinition = null)
{
Contracts.AssertValue(env);
Contracts.AssertValue(modelStream);
Contracts.AssertValueOrNull(inputSchemaDefinition);
Contracts.AssertValueOrNull(outputSchemaDefinition);
// Initialize pipe.
_srcDataView = DataViewConstructionUtils.CreateFromEnumerable(env, new TSrc[] { }, inputSchemaDefinition);
// Load transforms.
var pipe = env.LoadTransforms(modelStream, _srcDataView);
// Load predictor (if present) and apply default scorer.
// REVIEW: distinguish the case of predictor / no predictor?
var predictor = env.LoadPredictorOrNull(modelStream);
if (predictor != null)
{
var roles = ModelFileUtils.LoadRoleMappingsOrNull(env, modelStream);
pipe = roles != null
? env.CreateDefaultScorer(RoleMappedData.CreateOpt(pipe, roles), predictor)
: env.CreateDefaultScorer(env.CreateExamples(pipe, "Features"), predictor);
}
_pipeEngine = new PipeEngine <TDst>(env, pipe, ignoreMissingColumns, outputSchemaDefinition);
}
/// <summary>
/// Saves a model in a zip file.
/// </summary>
public static void SaveModel(IHostEnvironment env, IDataTransform tr, string outModelFilePath)
{
using (var ch = env.Start("SaveModel"))
using (var fs = File.Create(outModelFilePath))
{
var trainingExamples = env.CreateExamples(tr, null);
TrainUtils.SaveModel(env, ch, fs, null, trainingExamples);
}
}
/// <summary>
/// Constructor
/// </summary>
/// <param name="env">environment</param>
/// <param name="modelStream">stream</param>
/// <param name="output">name of the output column</param>
/// <param name="outputIsFloat">output is a gloat (true) or a vector of floats (false)</param>
/// <param name="conc">number of concurrency threads</param>
/// <param name="features">features name</param>
public ValueMapperPredictionEngineFloat(IHostEnvironment env, Stream modelStream,
string output = "Probability", bool outputIsFloat = true, int conc = 1,
string features = "Features")
{
_env = env;
if (_env == null)
{
throw Contracts.Except("env must not be null");
}
var inputs = new FloatVectorInput[0];
var view = ComponentCreation.CreateStreamingDataView <FloatVectorInput>(_env, inputs);
long modelPosition = modelStream.Position;
_predictor = ComponentCreation.LoadPredictorOrNull(_env, modelStream);
if (_predictor == null)
{
throw _env.Except("Unable to load a model.");
}
modelStream.Seek(modelPosition, SeekOrigin.Begin);
_transforms = ComponentCreation.LoadTransforms(_env, modelStream, view);
if (_transforms == null)
{
throw _env.Except("Unable to load a model.");
}
var data = _env.CreateExamples(_transforms, features);
if (data == null)
{
throw _env.Except("Cannot create rows.");
}
var scorer = _env.CreateDefaultScorer(data, _predictor);
if (scorer == null)
{
throw _env.Except("Cannot create a scorer.");
}
_valueMapper = new ValueMapperFromTransformFloat <VBuffer <float> >(_env,
scorer, features, output, conc: conc);
if (_valueMapper == null)
{
throw _env.Except("Cannot create a mapper.");
}
if (outputIsFloat)
{
_mapper = _valueMapper.GetMapper <VBuffer <float>, float>();
_mapperVector = null;
}
else
{
_mapper = null;
_mapperVector = _valueMapper.GetMapper <VBuffer <float>, VBuffer <float> >();
}
}
public ValueMapperExample(string modelName, string features)
{
_env = EnvHelper.NewTestEnvironment();
_predictor = ModelFileUtils.LoadPredictorOrNull(_env, File.OpenRead(modelName));
var inputs = new Input[0];
var view = DataViewConstructionUtils.CreateFromEnumerable(_env, inputs);
_transforms = ModelFileUtils.LoadTransforms(_env, view, File.OpenRead(modelName));
var data = _env.CreateExamples(_transforms, features);
var scorer = _env.CreateDefaultScorer(data, _predictor);
_valueMapper = new ValueMapperFromTransformFloat <VBuffer <float> >(_env, scorer, "Features", "Probability");
_mapper = _valueMapper.GetMapper <VBuffer <float>, float>();
}
public ValueMapperExample(string modelName, string features)
{
_env = EnvHelper.NewTestEnvironment();
_predictor = _env.LoadPredictorOrNull(File.OpenRead(modelName));
var inputs = new Input[0];
var view = _env.CreateStreamingDataView <Input>(inputs);
_transforms = ComponentCreation.LoadTransforms(_env, File.OpenRead(modelName), view);
var data = _env.CreateExamples(_transforms, features);
var scorer = _env.CreateDefaultScorer(data, _predictor);
_valueMapper = new ValueMapperFromTransformFloat <VBuffer <float> >(_env, scorer, "Features", "Probability");
_mapper = _valueMapper.GetMapper <VBuffer <float>, float>();
}
private static RoleMappedData CreateDataFromArgs(IHostEnvironment env, IExceptionContext ectx, IDataView input,
TrainAndScoreTransformer.ArgumentsBase args, out string feat, out string group)
{
var schema = input.Schema;
feat = TrainUtils.MatchNameOrDefaultOrNull(ectx, schema, "FeatureColumn", args.FeatureColumn, DefaultColumnNames.Features);
var label = TrainUtils.MatchNameOrDefaultOrNull(ectx, schema, "LabelColumn", args.LabelColumn, DefaultColumnNames.Label);
group = TrainUtils.MatchNameOrDefaultOrNull(ectx, schema, "GroupColumn", args.GroupColumn, DefaultColumnNames.GroupId);
var weight = TrainUtils.MatchNameOrDefaultOrNull(ectx, schema, "WeightColumn", args.WeightColumn, DefaultColumnNames.Weight);
var name = TrainUtils.MatchNameOrDefaultOrNull(ectx, schema, "NameColumn", args.NameColumn, DefaultColumnNames.Name);
var customCols_ = TrainUtils.CheckAndGenerateCustomColumns(ectx, args.CustomColumn);
var customCols = customCols_ == null ? new List <KeyValuePair <RoleMappedSchema.ColumnRole, string> >() : customCols_.ToList();
if (!string.IsNullOrEmpty(name))
{
customCols.Add(new KeyValuePair <RoleMappedSchema.ColumnRole, string>(RoleMappedSchema.ColumnRole.Name, name));
}
return(env.CreateExamples(input, feat, label: label, group: group, weight: weight, custom: customCols));
}
/// <summary>
/// Constructor
/// </summary>
/// <param name="env">environment</param>
/// <param name="modelStream">stream</param>
/// <param name="conc">number of concurrency threads</param>
/// <param name="features">features column</param>
public ValueMapperPredictionEngine(IHostEnvironment env, Stream modelStream,
int conc = 1, string features = "Features")
{
_env = env;
if (_env == null)
{
throw Contracts.Except("env must not be null");
}
var inputs = new TRowValue[0];
var view = ComponentCreation.CreateStreamingDataView <TRowValue>(_env, inputs);
long modelPosition = modelStream.Position;
_predictor = ComponentCreation.LoadPredictorOrNull(_env, modelStream);
if (_predictor == null)
{
throw _env.Except("Unable to load a model.");
}
modelStream.Seek(modelPosition, SeekOrigin.Begin);
_transforms = ComponentCreation.LoadTransforms(_env, modelStream, view);
if (_transforms == null)
{
throw _env.Except("Unable to load a model.");
}
var data = _env.CreateExamples(_transforms, features);
if (data == null)
{
throw _env.Except("Cannot create rows.");
}
var scorer = _env.CreateDefaultScorer(data, _predictor);
if (scorer == null)
{
throw _env.Except("Cannot create a scorer.");
}
_CreateMapper(scorer, conc);
}
/// <summary>
/// Finalize the test on a transform, calls the transform,
/// saves the data, saves the models, loads it back, saves the data again,
/// checks the output is the same.
/// </summary>
/// <param name="env">environment</param>
/// <param name="outModelFilePath"model filename</param>
/// <param name="transform">transform to test</param>
/// <param name="source">source (view before applying the transform</param>
/// <param name="outData">fist data file</param>
/// <param name="outData2">second data file</param>
/// <param name="startsWith">Check that outputs is the same on disk after outputting the transformed data after the model was serialized</param>
public static void SerializationTestTransform(IHostEnvironment env,
string outModelFilePath, IDataTransform transform,
IDataView source, string outData, string outData2,
bool startsWith = false, bool skipDoubleQuote = false,
bool forceDense = false)
{
// Saves model.
var roles = env.CreateExamples(transform, null);
using (var ch = env.Start("SaveModel"))
using (var fs = File.Create(outModelFilePath))
TrainUtils.SaveModel(env, ch, fs, null, roles);
if (!File.Exists(outModelFilePath))
{
throw new FileNotFoundException(outModelFilePath);
}
// We load it again.
using (var fs = File.OpenRead(outModelFilePath))
{
var tr2 = env.LoadTransforms(fs, source);
if (tr2 == null)
{
throw new Exception(string.Format("Unable to load '{0}'", outModelFilePath));
}
if (transform.GetType() != tr2.GetType())
{
throw new Exception(string.Format("Type mismatch {0} != {1}", transform.GetType(), tr2.GetType()));
}
}
// Checks the outputs.
var saver = env.CreateSaver(forceDense ? "Text{dense=+}" : "Text");
var columns = new int[transform.Schema.Count];
for (int i = 0; i < columns.Length; ++i)
{
columns[i] = i;
}
using (var fs2 = File.Create(outData))
saver.SaveData(fs2, transform, columns);
if (!File.Exists(outModelFilePath))
{
throw new FileNotFoundException(outData);
}
// Check we have the same output.
using (var fs = File.OpenRead(outModelFilePath))
{
var tr = env.LoadTransforms(fs, source);
saver = env.CreateSaver(forceDense ? "Text{dense=+}" : "Text");
using (var fs2 = File.Create(outData2))
saver.SaveData(fs2, tr, columns);
}
var t1 = File.ReadAllLines(outData);
var t2 = File.ReadAllLines(outData2);
if (t1.Length != t2.Length)
{
throw new Exception(string.Format("Not the same number of lines: {0} != {1}", t1.Length, t2.Length));
}
for (int i = 0; i < t1.Length; ++i)
{
if (skipDoubleQuote && (t1[i].Contains("\"\"\t\"\"") || t2[i].Contains("\"\"\t\"\"")))
{
continue;
}
if ((startsWith && !t1[i].StartsWith(t2[i])) || (!startsWith && t1[i] != t2[i]))
{
if (t1[i].EndsWith("\t5\t0:\"\""))
{
var a = t1[i].Substring(0, t1[i].Length - "\t5\t0:\"\"".Length);
a += "\t\"\"\t\"\"\t\"\"\t\"\"\t\"\"";
var b = t2[i];
if ((startsWith && !a.StartsWith(b)) || (!startsWith && a != b))
{
throw new Exception(string.Format("2-Mismatch on line {0}/{3}:\n{1}\n{2}", i, t1[i], t2[i], t1.Length));
}
}
else
{
// The test might fail because one side is dense and the other is sparse.
throw new Exception(string.Format("3-Mismatch on line {0}/{3}:\n{1}\n{2}", i, t1[i], t2[i], t1.Length));
}
}
}
}
