public TransposeLoader(IHostEnvironment env, Arguments args, IMultiStreamSource file)
{
Contracts.CheckValue(env, nameof(env));
_host = env.Register(LoadName);
_host.CheckValue(args, nameof(args));
_host.CheckValue(file, nameof(file));
_host.Check(file.Count == 1, "Transposed loader accepts a single file only");
_threads = args.Threads ?? 0;
if (_threads < 0)
{
_threads = 0;
}
_file = file;
using (Stream stream = _file.Open(0))
using (BinaryReader reader = new BinaryReader(stream))
{
_header = InitHeader(reader);
reader.Seek(_header.SubIdvTableOffset);
_schemaEntry = new SubIdvEntry.SchemaSubIdv(this, reader);
_entries = new SubIdvEntry.TransposedSubIdv[_header.ColumnCount];
for (int c = 0; c < _entries.Length; ++c)
{
_entries[c] = new SubIdvEntry.TransposedSubIdv(this, reader, c);
}
_schema = new SchemaImpl(this);
if (!HasRowData)
{
_colTransposers = new Transposer[_header.ColumnCount];
_colTransposersLock = new object();
}
}
}
public ResampleTransform(IHostEnvironment env, Arguments args, IDataView input)
{
Contracts.CheckValue(env, "env");
_host = env.Register("ResampleTransform");
_host.CheckValue(args, "args"); // Checks values are valid.
_host.CheckValue(input, "input");
_host.Check(args.lambda > 0, "lambda must be > 0");
_input = input;
_args = args;
_cacheReplica = null;
if (!string.IsNullOrEmpty(_args.column))
{
int index;
if (!_input.Schema.TryGetColumnIndex(_args.column, out index))
{
throw _host.Except("Unable to find column '{0}' in\n'{1}'.", _args.column, SchemaHelper.ToString(_input.Schema));
}
if (string.IsNullOrEmpty(_args.classValue))
{
throw _host.Except("Class value cannot be null.");
}
}
_transform = CreateTemplatedTransform();
}
public void Read(ModelLoadContext ctx, IHost host)
{
string sr = ctx.Reader.ReadString();
host.CheckValue(sr, "columns");
columns = sr.Split(',').Where(c => !string.IsNullOrEmpty(c)).ToArray();
if (columns.Length == 0)
{
columns = null;
}
sr = ctx.Reader.ReadString();
host.CheckValue(sr, "hists");
hists = sr.Split(',').Where(c => !string.IsNullOrEmpty(c)).ToArray();
if (hists.Length == 0)
{
hists = null;
}
saveInFile = ctx.Reader.ReadString();
int nb = ctx.Reader.ReadInt32();
host.Check(nb == 0 || nb == 1, "passThrough");
passThrough = nb == 1;
showSchema = ctx.Reader.ReadInt32() == 1;
dimension = ctx.Reader.ReadInt32() == 1;
oneRowPerColumn = ctx.Reader.ReadInt32() == 1;
jsonFormat = ctx.Reader.ReadInt32() == 1;
name = ctx.Reader.ReadString();
}
private TransposeLoader(IHost host, ModelLoadContext ctx, IMultiStreamSource file)
{
Contracts.CheckValue(host, nameof(host));
_host = host;
_host.CheckValue(file, nameof(file));
_host.Check(file.Count == 1, "Transposed loader accepts a single file only");
// *** Binary format **
// int: Number of threads if explicitly defined, or 0 if the
// number of threads was automatically determined
_threads = ctx.Reader.ReadInt32();
_host.CheckDecode(_threads >= 0);
// Dedupe code somehow?
_file = file;
using (Stream stream = _file.Open(0))
using (BinaryReader reader = new BinaryReader(stream))
{
_header = InitHeader(reader);
reader.Seek(_header.SubIdvTableOffset);
_schemaEntry = new SubIdvEntry.SchemaSubIdv(this, reader);
_entries = new SubIdvEntry.TransposedSubIdv[_header.ColumnCount];
for (int c = 0; c < _entries.Length; ++c)
{
_entries[c] = new SubIdvEntry.TransposedSubIdv(this, reader, c);
}
if (!HasRowData)
{
_colTransposers = new Transposer[_header.ColumnCount];
_colTransposersLock = new object();
}
}
}
protected string GetHashKey(long transformsBitMask, RecipeInference.SuggestedRecipe.SuggestedLearner learner)
{
var learnerName = learner.ToString();
Host.Check(!string.IsNullOrEmpty(learnerName));
return($"{learnerName}+{transformsBitMask}");
}
void ICanSaveModel.Save(ModelSaveContext ctx)
{
Host.Check(Meta != null, "Can't save an untrained Stacking combiner");
Host.CheckValue(ctx, nameof(ctx));
ctx.CheckAtModel();
SaveCore(ctx);
}
internal DateTimeEstimator(IHostEnvironment env, Options options)
{
Contracts.CheckValue(env, nameof(env));
_host = Contracts.CheckRef(env, nameof(env)).Register("DateTimeTransformerEstimator");
_host.Check(!CommonExtensions.OsIsCentOS7(), "CentOS7 is not supported");
_options = options;
}
IPredictor ITrainer <IPredictor> .Train(TrainContext context)
{
Host.CheckValue(context, nameof(context));
var pred = TrainModelCore(context) as IPredictor;
Host.Check(pred != null, "Training did not return a predictor.");
return(pred);
}
private static void TransformFeatures(IHost host, ref VBuffer <Float> src, ref VBuffer <Float> dst, TransformInfo transformInfo,
AlignedArray featuresAligned, AlignedArray productAligned)
{
Contracts.AssertValue(host, "host");
host.Check(src.Length == transformInfo.SrcDim, "column does not have the expected dimensionality.");
var values = dst.Values;
Float scale;
if (transformInfo.RotationTerms != null)
{
if (Utils.Size(values) < transformInfo.NewDim)
{
values = new Float[transformInfo.NewDim];
}
scale = MathUtils.Sqrt((Float)2.0 / transformInfo.NewDim);
}
else
{
if (Utils.Size(values) < 2 * transformInfo.NewDim)
{
values = new Float[2 * transformInfo.NewDim];
}
scale = MathUtils.Sqrt((Float)1.0 / transformInfo.NewDim);
}
if (src.IsDense)
{
featuresAligned.CopyFrom(src.Values, 0, src.Length);
CpuMathUtils.MatTimesSrc(false, false, transformInfo.RndFourierVectors, featuresAligned, productAligned,
transformInfo.NewDim);
}
else
{
// This overload of MatTimesSrc ignores the values in slots that are not in src.Indices, so there is
// no need to zero them out.
featuresAligned.CopyFrom(src.Indices, src.Values, 0, 0, src.Count, zeroItems: false);
CpuMathUtils.MatTimesSrc(false, false, transformInfo.RndFourierVectors, src.Indices, featuresAligned, 0, 0,
src.Count, productAligned, transformInfo.NewDim);
}
for (int i = 0; i < transformInfo.NewDim; i++)
{
var dotProduct = productAligned[i];
if (transformInfo.RotationTerms != null)
{
values[i] = (Float)MathUtils.Cos(dotProduct + transformInfo.RotationTerms[i]) * scale;
}
else
{
values[2 * i] = (Float)MathUtils.Cos(dotProduct) * scale;
values[2 * i + 1] = (Float)MathUtils.Sin(dotProduct) * scale;
}
}
dst = new VBuffer <Float>(transformInfo.RotationTerms == null ? 2 * transformInfo.NewDim : transformInfo.NewDim,
values, dst.Indices);
}
IRowToRowMapper ITransformer.GetRowToRowMapper(DataViewSchema inputSchema)
{
_host.CheckValue(inputSchema, nameof(inputSchema));
var transform = ApplyTransformUtils.ApplyTransformToData(_host, (IDataTransform)_xf, new EmptyDataView(_host, inputSchema)) as IRowToRowMapper;
_host.Check(transform is IRowToRowMapper);
return(new CompositeRowToRowMapper(inputSchema, new[] { transform }));
}
private void PrepareNextBatch(IEnumerable <IRunResult> results)
{
_host.Check(!_disposed, "Creating parameters while sweeper is disposed");
var paramSets = _baseSweeper.ProposeSweeps(_batchSize, results);
if (Utils.Size(paramSets) == 0)
{
// Mark the queue as completed.
_paramChannel.Writer.Complete();
return;
}
// Assign an id to each ParameterSet and enque it.
foreach (var paramSet in paramSets)
{
_paramChannel.Writer.TryWrite(new ParameterSetWithId(_numGenerated++, paramSet));
}
EnsureResultsSize();
}
public void SaveCore(ModelSaveContext ctx, IHost host, VersionInfo versionInfo)
{
host.Check(Classes.Count > 0, "The model cannot be saved, it was never trained.");
host.Check(Classes.Count == Classes.Length, "The model cannot be saved, it was never trained.");
ctx.SetVersionInfo(versionInfo);
ctx.Writer.WriteIntArray(Classes.Indices);
if (LabelType == NumberDataViewType.Single)
{
ctx.Writer.WriteSingleArray(Classes.Values as float[]);
}
else if (LabelType == NumberDataViewType.Byte)
{
ctx.Writer.WriteByteArray(Classes.Values as byte[]);
}
else if (LabelType == NumberDataViewType.UInt16)
{
ctx.Writer.WriteUIntArray((Classes.Values as ushort[]).Select(c => (uint)c).ToArray());
}
else if (LabelType == NumberDataViewType.UInt32)
{
ctx.Writer.WriteUIntArray(Classes.Values as uint[]);
}
else
{
throw host.Except("Unexpected type for LabelType.");
}
ctx.Writer.Write(_singleColumn ? 1 : 0);
ctx.Writer.Write(_labelKey ? 1 : 0);
var preds = Predictors;
ctx.Writer.Write(preds.Length);
for (int i = 0; i < preds.Length; i++)
{
ctx.SaveModel(preds[i], string.Format("M2B{0}", i));
}
ctx.Writer.Write(_reclassificationPredictor != null ? (byte)1 : (byte)0);
if (_reclassificationPredictor != null)
{
ctx.SaveModel(_reclassificationPredictor, "Reclassification");
}
ctx.Writer.Write((byte)213);
}
/// <summary>
/// Constructor.
/// </summary>
/// <param name="env">environment</param>
/// <param name="input">input source stored as the secondary source</param>
/// <param name="name">name of the transform</param>
public AbstractSimpleTransformTemplate(IHostEnvironment env, IDataView input, string name)
{
Contracts.CheckValue(env, "env");
_host = env.Register(name);
_host.CheckValue(input, "input");
_host.Check(!string.IsNullOrEmpty(name));
_sourceCtx = input;
_sourcePipe = null;
_lock = new object();
}
public unsafe void Get(out int m, out int n, out int k, out float[] p, out float[] q)
{
_host.Check(_pMFModel != null, "Attempted to get predictor before training");
m = _pMFModel->M;
_host.Check(m > 0, "Number of rows should have been positive but was not");
n = _pMFModel->N;
_host.Check(n > 0, "Number of columns should have been positive but was not");
k = _pMFModel->K;
_host.Check(k > 0, "Internal dimension should have been positive but was not");
p = new float[m * k];
q = new float[n * k];
unsafe
{
Marshal.Copy((IntPtr)_pMFModel->P, p, 0, p.Length);
Marshal.Copy((IntPtr)_pMFModel->Q, q, 0, q.Length);
}
}
public RandomFeatureSelector(IHostEnvironment env, Arguments args)
{
Contracts.CheckValue(env, nameof(env));
env.CheckValue(args, nameof(args));
_host = env.Register(LoadName);
_args = args;
_host.Check(0 < _args.FeaturesSelectionProportion && _args.FeaturesSelectionProportion < 1,
"The feature proportion for RandomFeatureSelector should be greater than 0 and lesser than 1");
}
internal ColumnSelectingTransformer(IHostEnvironment env, string[] keepColumns, string[] dropColumns,
bool keepHidden = ColumnSelectingEstimator.Defaults.KeepHidden, bool ignoreMissing = ColumnSelectingEstimator.Defaults.IgnoreMissing)
{
_host = Contracts.CheckRef(env, nameof(env)).Register(nameof(ColumnSelectingTransformer));
_host.CheckValueOrNull(keepColumns);
_host.CheckValueOrNull(dropColumns);
bool keepValid = Utils.Size(keepColumns) > 0;
bool dropValid = Utils.Size(dropColumns) > 0;
// Check that both are not valid
_host.Check(!(keepValid && dropValid), "Both " + nameof(keepColumns) + " and " + nameof(dropColumns) + " are set. Exactly one can be specified.");
// Check that both are invalid
_host.Check(!(!keepValid && !dropValid), "Neither " + nameof(keepColumns) + " and " + nameof(dropColumns) + " is set. Exactly one must be specified.");
_selectedColumns = (keepValid) ? keepColumns : dropColumns;
KeepColumns = keepValid;
KeepHidden = keepHidden;
IgnoreMissing = ignoreMissing;
}
/// <summary>
/// Loading constructor.
/// </summary>
/// <param name="ctx">reading context</param>
/// <param name="env">environment</param>
/// <param name="input">input source stored as the secondary source</param>
/// <param name="name">name of the transform</param>
public AbstractSimpleTransformTemplate(IHost host, ModelLoadContext ctx, IDataView input, string name)
{
Contracts.CheckValue(host, "env");
_host = host;
_host.CheckValue(ctx, "env");
_host.CheckValue(input, "input");
_host.Check(!string.IsNullOrEmpty(name));
_sourceCtx = input;
_sourcePipe = null;
_lock = new object();
}
public SelectColumnsTransform(IHostEnvironment env, string[] keepColumns, string[] dropColumns,
bool keepHidden = true, bool ignoreMissing = true)
{
_host = Contracts.CheckRef(env, nameof(env)).Register(nameof(SelectColumnsTransform));
_host.CheckValueOrNull(keepColumns);
_host.CheckValueOrNull(dropColumns);
bool keepValid = keepColumns != null && keepColumns.Count() > 0;
bool dropValid = dropColumns != null && dropColumns.Count() > 0;
// Check that both are not valid
_host.Check(!(keepValid && dropValid), "Both keepColumns and dropColumns are set, only one can be specified.");
// Check that both are invalid
_host.Check(!(!keepValid && !dropValid), "Neither keepColumns or dropColumns is set, one must be specified.");
_selectedColumns = (keepValid) ? keepColumns : dropColumns;
KeepColumns = keepValid;
KeepHidden = keepHidden;
IgnoreMissing = ignoreMissing;
}
public TransformApplierParams(IHost host, Arguments args)
{
Contracts.AssertValue(host);
host.CheckUserArg(args.Column != null, nameof(args.Column), "Columns must be specified");
host.CheckUserArg(args.WordFeatureExtractor != null || args.CharFeatureExtractor != null || args.OutputTokens,
nameof(args.WordFeatureExtractor), "At least one feature extractor or OutputTokens must be specified.");
host.Check(Enum.IsDefined(typeof(Language), args.Language));
host.Check(Enum.IsDefined(typeof(CaseNormalizationMode), args.TextCase));
WordExtractorFactory = args.WordFeatureExtractor?.CreateComponent(host, args.Dictionary);
CharExtractorFactory = args.CharFeatureExtractor?.CreateComponent(host, args.Dictionary);
VectorNormalizer = args.VectorNormalizer;
Language = args.Language;
StopWordsRemover = args.StopWordsRemover;
TextCase = args.TextCase;
KeepDiacritics = args.KeepDiacritics;
KeepPunctuations = args.KeepPunctuations;
KeepNumbers = args.KeepNumbers;
OutputTextTokens = args.OutputTokens;
Dictionary = args.Dictionary;
}
internal CountTargetEncodingEstimator(IHostEnvironment env, string labelColumnName, CountTargetEncodingTransformer initialCounts, params InputOutputColumnPair[] columns)
{
Contracts.CheckValue(env, nameof(env));
_host = env.Register(nameof(CountTargetEncodingEstimator));
_host.CheckValue(initialCounts, nameof(initialCounts));
_host.CheckNonEmpty(columns, nameof(columns));
_host.Check(initialCounts.VerifyColumns(columns), nameof(columns));
_hashingEstimator = new HashingEstimator(_host, initialCounts.HashingTransformer.Columns.ToArray());
_countTableEstimator = new CountTableEstimator(_host, labelColumnName, initialCounts.CountTable,
columns.Select(c => new InputOutputColumnPair(c.OutputColumnName, c.OutputColumnName)).ToArray());
}
private SchemaBindablePipelineEnsembleBase(IHostEnvironment env, IPredictorModel[] predictors, string registrationName, string scoreColumnKind)
{
Contracts.CheckValue(env, nameof(env));
Host = env.Register(registrationName);
Host.CheckNonEmpty(predictors, nameof(predictors));
Host.CheckNonWhiteSpace(scoreColumnKind, nameof(scoreColumnKind));
PredictorModels = predictors;
_scoreColumnKind = scoreColumnKind;
HashSet <string> inputCols = null;
for (int i = 0; i < predictors.Length; i++)
{
var predModel = predictors[i];
// Get the input column names.
var inputSchema = predModel.TransformModel.InputSchema;
if (inputCols == null)
{
inputCols = new HashSet <string>();
for (int j = 0; j < inputSchema.ColumnCount; j++)
{
if (inputSchema.IsHidden(j))
{
continue;
}
inputCols.Add(inputSchema.GetColumnName(j));
}
_inputCols = inputCols.ToArray();
}
else
{
int nonHiddenCols = 0;
for (int j = 0; j < inputSchema.ColumnCount; j++)
{
if (inputSchema.IsHidden(j))
{
continue;
}
var name = inputSchema.GetColumnName(j);
if (!inputCols.Contains(name))
{
throw Host.Except("Inconsistent schemas: Some schemas do not contain the column '{0}'", name);
}
nonHiddenCols++;
}
Host.Check(nonHiddenCols == _inputCols.Length,
"Inconsistent schemas: not all schemas have the same number of columns");
}
}
}
internal TimeSeriesImputerEstimator(IHostEnvironment env, string timeSeriesColumn, string[] grainColumns, string[] filterColumns, FilterMode filterMode, ImputationStrategy imputeMode, bool supressTypeErrors)
{
Contracts.CheckValue(env, nameof(env));
_host.Check(!CommonExtensions.OsIsCentOS7(), "CentOS7 is not supported");
_host = Contracts.CheckRef(env, nameof(env)).Register("TimeSeriesImputerEstimator");
_host.CheckValue(timeSeriesColumn, nameof(timeSeriesColumn), "TimePoint column should not be null.");
_host.CheckNonEmpty(grainColumns, nameof(grainColumns), "Need at least one grain column.");
if (filterMode == FilterMode.Include)
{
_host.CheckNonEmpty(filterColumns, nameof(filterColumns), "Need at least 1 filter column if a FilterMode is specified");
}
_options = new Options
{
TimeSeriesColumn = timeSeriesColumn,
GrainColumns = grainColumns,
FilterColumns = filterColumns == null ? new string[] { } : filterColumns,
FilterMode = filterMode,
ImputeMode = imputeMode,
SupressTypeErrors = supressTypeErrors
};
}
internal CategoricalImputerEstimator(IHostEnvironment env, Options options)
{
Contracts.CheckValue(env, nameof(env));
_host.Check(!CommonExtensions.OsIsCentOS7(), "CentOS7 is not supported");
_host = env.Register(nameof(CategoricalImputerEstimator));
foreach (var columnPair in options.Columns)
{
columnPair.Source = columnPair.Source ?? columnPair.Name;
}
_options = options;
}
private void SetInput(Schema schema, HashSet<string> toDrop)
{
var recordType = new JObject();
recordType["type"] = "record";
recordType["name"] = "DataInput";
var fields = new JArray();
var fieldNames = new HashSet<string>();
for (int c = 0; c < schema.Count; ++c)
{
if (schema[c].IsHidden)
continue;
string name = schema[c].Name;
if (toDrop.Contains(name))
continue;
JToken pfaType = PfaTypeOrNullForColumn(schema, c);
if (pfaType == null)
continue;
string fieldName = ModelUtils.CreateNameCore(name, fieldNames.Contains);
fieldNames.Add(fieldName);
var fieldDeclaration = new JObject();
fieldDeclaration["name"] = fieldName;
fieldDeclaration["type"] = pfaType;
fields.Add(fieldDeclaration);
_nameToVarName.Add(name, "input." + fieldName);
}
_host.Assert(_nameToVarName.Count == fields.Count);
_host.Assert(_nameToVarName.Count == fieldNames.Count);
recordType["fields"] = fields;
_host.Check(fields.Count >= 1, "Schema produced no inputs for the PFA conversion.");
if (fields.Count == 1)
{
// If there's only one, don't bother forming a record.
var field = (JObject)fields[0];
Pfa.InputType = field["type"];
_nameToVarName[_nameToVarName.Keys.First()] = "input";
}
else
Pfa.InputType = recordType;
}
public override string AddInitializer(IEnumerable <bool> values, IEnumerable <long> dims, string name = null, bool makeUniqueName = true)
{
_host.CheckValue(values, nameof(values));
if (dims != null)
{
_host.Check(dims.Aggregate((x, y) => x * y) == values.Count(), "Number of elements doesn't match tensor size");
}
name = AddVariable(name ?? "bools", makeUniqueName);
_initializers.Add(OnnxUtils.MakeInt32s(name, typeof(bool), values.Select(v => Convert.ToInt32(v)), dims));
return(name);
}
internal DateTimeEstimator(IHostEnvironment env, string inputColumnName, string columnPrefix, HolidayList country = HolidayList.None)
{
Contracts.CheckValue(env, nameof(env));
_host = Contracts.CheckRef(env, nameof(env)).Register("DateTimeTransformerEstimator");
_host.CheckValue(inputColumnName, nameof(inputColumnName), "Input column should not be null.");
_host.Check(!CommonExtensions.OsIsCentOS7(), "CentOS7 is not supported");
_options = new Options
{
Source = inputColumnName,
Prefix = columnPrefix,
Country = country
};
}
/// <summary>
/// Save model to the given context
/// </summary>
public void Save(ModelSaveContext ctx)
{
ctx.CheckAtModel();
ctx.SetVersionInfo(GetVersionInfo());
// *** Binary format ***
// int: number of rows (m), the limit on row
// int: number of columns (n), the limit on column
// int: rank of factor matrices (k)
// float[m * k]: the left factor matrix
// float[k * n]: the right factor matrix
_host.Check(_numberOfRows > 0, "Number of rows must be positive");
_host.Check(_numberofColumns > 0, "Number of columns must be positive");
_host.Check(_approximationRank > 0, "Number of latent factors must be positive");
ctx.Writer.Write(_numberOfRows);
ctx.Writer.Write(_numberofColumns);
ctx.Writer.Write(_approximationRank);
_host.Check(Utils.Size(_leftFactorMatrix) == _numberOfRows * _approximationRank, "Unexpected matrix size of a factor matrix (matrix P in LIBMF paper)");
_host.Check(Utils.Size(_rightFactorMatrix) == _numberofColumns * _approximationRank, "Unexpected matrix size of a factor matrix (matrix Q in LIBMF paper)");
Utils.WriteSinglesNoCount(ctx.Writer, _leftFactorMatrix.AsSpan(0, _numberOfRows * _approximationRank));
Utils.WriteSinglesNoCount(ctx.Writer, _rightFactorMatrix.AsSpan(0, _numberofColumns * _approximationRank));
}
/// <summary>
/// Returns the feature selection scores for each slot of each column.
/// </summary>
/// <param name="host">The host.</param>
/// <param name="input">The input dataview.</param>
/// <param name="labelColumnName">The label column.</param>
/// <param name="columns">The columns for which to compute the feature selection scores.</param>
/// <param name="numBins">The number of bins to use for numeric features.</param>
/// <returns>A list of scores for each column and each slot.</returns>
public static Single[][] Train(IHost host, IDataView input, string labelColumnName, string[] columns, int numBins)
{
Contracts.CheckValue(host, nameof(host));
host.CheckValue(input, nameof(input));
host.CheckNonWhiteSpace(labelColumnName, nameof(labelColumnName));
host.CheckValue(columns, nameof(columns));
host.Check(columns.Length > 0, "At least one column must be specified.");
host.Check(numBins > 1, "numBins must be greater than 1.");
HashSet <string> colSet = new HashSet <string>();
foreach (string col in columns)
{
if (!colSet.Add(col))
{
throw host.Except("Column '{0}' specified multiple times.", col);
}
}
var colSizes = new int[columns.Length];
return(TrainCore(host, input, labelColumnName, columns, numBins, colSizes));
}
public BagMultiCountTableBuilder(IHostEnvironment env, MultiCountTable table, DataViewSchema.Column[] inputCols, long labelCardinality)
{
Contracts.CheckValue(env, nameof(env));
env.CheckValue(table, nameof(table));
_host = env.Register(LoaderSignature);
_host.Check(Utils.Size(inputCols) == table.ColCount, "Inconsistent number of columns");
_host.Check(table.SlotCount.Zip(inputCols, (count, col) => (count, col)).
All(pair => pair.col.Type.GetValueCount() == pair.count), "Inconsistent number of slots");
_builder = table.BaseTable.ToBuilder(labelCardinality);
_colCount = table.ColCount;
_slotCount = new int[_colCount];
table.SlotCount.CopyTo(_slotCount, 0);
}
private Delegate CreateKeyGetterDelegate <TDst>(Delegate peekDel, ColumnType colType)
{
// Make sure the function is dealing with key.
Host.Check(colType.IsKey);
// Following equations work only with contiguous key type.
Host.Check(colType.AsKey.Contiguous);
// Following equations work only with unsigned integers.
Host.Check(typeof(TDst) == typeof(ulong) || typeof(TDst) == typeof(uint) ||
typeof(TDst) == typeof(byte) || typeof(TDst) == typeof(bool));
// Convert delegate function to a function which can fetch the underlying value.
var peek = peekDel as Peek <TRow, TDst>;
Host.AssertValue(peek);
TDst rawKeyValue = default;
ulong key = 0; // the raw key value as ulong
ulong min = colType.AsKey.Min;
ulong max = min + (ulong)colType.AsKey.Count - 1;
ulong result = 0; // the result as ulong
ValueGetter <TDst> getter = (ref TDst dst) =>
{
peek(GetCurrentRowObject(), Position, ref rawKeyValue);
key = (ulong)Convert.ChangeType(rawKeyValue, typeof(ulong));
if (min <= key && key <= max)
{
result = key - min + 1;
}
else
{
result = 0;
}
dst = (TDst)Convert.ChangeType(result, typeof(TDst));
};
return(getter);
}