/// <summary>
/// Load a <see cref="IDataView"/> from a text file using <see cref="TextLoader"/>.
/// Note that <see cref="IDataView"/>'s are lazy, so no actual loading happens here, just schema validation.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="path">The path to the file(s).</param>
/// <param name="columns">The columns of the schema.</param>
/// <param name="separatorChar">The character used as separator between data points in a row. By default the tab character is used as separator.</param>
/// <param name="hasHeader">Whether the file has a header. When <see langword="true"/>, the loader will skip the first line when
/// <see cref="TextLoader.Load(IMultiStreamSource)"/> is called.</param>
/// <param name="allowQuoting">Whether the input may include double-quoted values. This parameter is used to distinguish separator characters
/// in an input value from actual separators. When <see langword="true"/>, separators within double quotes are treated as part of the
/// input value. When <see langword="false"/>, all separators, even those whitin quotes, are treated as delimiting a new column.
/// It is also used to distinguish empty values from missing values. When <see langword="true"/>, missing value are denoted by consecutive
/// separators and empty values by \"\". When <see langword="false"/>, empty values are denoted by consecutive separators and missing
/// values by the default missing value for each type documented in <see cref="DataKind"/>.</param>
/// <param name="trimWhitespace">Remove trailing whitespace from lines.</param>
/// <param name="allowSparse">Whether the input may include sparse representations. For example, a row containing
/// "5 2:6 4:3" means that there are 5 columns, and the only non-zero are columns 2 and 4, which have values 6 and 3,
/// respectively. Column indices are zero-based, so columns 2 and 4 represent the 3rd and 5th columns.
/// A column may also have dense values followed by sparse values represented in this fashion. For example,
/// a row containing "1 2 5 2:6 4:3" represents two dense columns with values 1 and 2, followed by 5 sparsely represented
/// columns with values 0, 0, 6, 0, and 3. The indices of the sparse columns start from 0, even though 0 represents the third column.</param>
/// <returns>The data view.</returns>
public static IDataView LoadFromTextFile(this DataOperationsCatalog catalog,
string path,
TextLoader.Column[] columns,
char separatorChar = TextLoader.Defaults.Separator,
bool hasHeader = TextLoader.Defaults.HasHeader,
bool allowQuoting = TextLoader.Defaults.AllowQuoting,
bool trimWhitespace = TextLoader.Defaults.TrimWhitespace,
bool allowSparse = TextLoader.Defaults.AllowSparse)
{
CheckValidPathContents(path);
var options = new TextLoader.Options
{
Columns = columns,
Separators = new[] { separatorChar },
HasHeader = hasHeader,
AllowQuoting = allowQuoting,
TrimWhitespace = trimWhitespace,
AllowSparse = allowSparse
};
var loader = new TextLoader(CatalogUtils.GetEnvironment(catalog), options: options);
return(loader.Load(new MultiFileSource(path)));
}
/// <summary>
/// Create a text loader <see cref="TextLoader"/> by inferencing the dataset schema from a data model type.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="hasHeader">Does the file contains header?</param>
/// <param name="separatorChar">Column separator character. Default is '\t'</param>
/// <param name="allowQuotedStrings">Whether the input may include quoted values,
/// which can contain separator characters, colons,
/// and distinguish empty values from missing values. When true, consecutive separators
/// denote a missing value and an empty value is denoted by \"\".
/// When false, consecutive separators denote an empty value.</param>
/// <param name="supportSparse">Whether the input may include sparse representations for example,
/// if one of the row contains "5 2:6 4:3" that's mean there are 5 columns all zero
/// except for 3rd and 5th columns which have values 6 and 3</param>
/// <param name="trimWhitespace">Remove trailing whitespace from lines</param>
public static TextLoader CreateTextLoader <TInput>(this DataOperationsCatalog catalog,
bool hasHeader = TextLoader.DefaultArguments.HasHeader,
char separatorChar = TextLoader.DefaultArguments.Separator,
bool allowQuotedStrings = TextLoader.DefaultArguments.AllowQuoting,
bool supportSparse = TextLoader.DefaultArguments.AllowSparse,
bool trimWhitespace = TextLoader.DefaultArguments.TrimWhitespace)
=> TextLoader.CreateTextReader <TInput>(CatalogUtils.GetEnvironment(catalog), hasHeader, separatorChar, allowQuotedStrings, supportSparse, trimWhitespace);
private protected CrossValidationResult[] CrossValidateTrain(IDataView data, IEstimator <ITransformer> estimator,
int numFolds, string samplingKeyColumn, int?seed = null)
{
Environment.CheckValue(data, nameof(data));
Environment.CheckValue(estimator, nameof(estimator));
Environment.CheckParam(numFolds > 1, nameof(numFolds), "Must be more than 1");
Environment.CheckValueOrNull(samplingKeyColumn);
DataOperationsCatalog.EnsureGroupPreservationColumn(Environment, ref data, ref samplingKeyColumn, seed);
var result = new CrossValidationResult[numFolds];
int fold = 0;
// Sequential per-fold training.
// REVIEW: we could have a parallel implementation here. We would need to
// spawn off a separate host per fold in that case.
foreach (var split in DataOperationsCatalog.CrossValidationSplit(Environment, data, numFolds, samplingKeyColumn))
{
var model = estimator.Fit(split.TrainSet);
var scoredTest = model.Transform(split.TestSet);
result[fold] = new CrossValidationResult(model, scoredTest, fold);
fold++;
}
return(result);
}
/// <summary>
/// Load a <see cref="IDataView"/> from a text file using <see cref="TextLoader"/>.
/// Note that <see cref="IDataView"/>'s are lazy, so no actual loading happens here, just schema validation.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="path">The path to the file.</param>
/// <param name="columns">The columns of the schema.</param>
/// <param name="separatorChar">The character used as separator between data points in a row. By default the tab character is used as separator.</param>
/// <param name="hasHeader">Whether the file has a header with feature names. Note: If a TextLoader is created with hasHeader = true but without a
/// dataSample, then vector columns made by TextLoader will not contain slot name annotations (slots being the elements of the given vector column),
/// because the output schema is made when the TextLoader is made, and not when <see cref="TextLoader.Load(IMultiStreamSource)"/> is called.
/// In addition, the case where dataSample = null and hasHeader = true indicates to the loader that when it is given a file when Load()
/// is called, it needs to skip the first line.</param>
/// <param name="allowQuoting">Whether the file can contain columns defined by a quoted string.</param>
/// <param name="trimWhitespace">Remove trailing whitespace from lines</param>
/// <param name="allowSparse">Whether the file can contain numerical vectors in sparse format.</param>
/// <returns>The data view.</returns>
public static IDataView LoadFromTextFile(this DataOperationsCatalog catalog,
string path,
TextLoader.Column[] columns,
char separatorChar = TextLoader.Defaults.Separator,
bool hasHeader = TextLoader.Defaults.HasHeader,
bool allowQuoting = TextLoader.Defaults.AllowQuoting,
bool trimWhitespace = TextLoader.Defaults.TrimWhitespace,
bool allowSparse = TextLoader.Defaults.AllowSparse)
{
Contracts.CheckNonEmpty(path, nameof(path));
if (!File.Exists(path))
{
throw Contracts.ExceptParam(nameof(path), "File does not exist at path: {0}", path);
}
var options = new TextLoader.Options
{
Columns = columns,
Separators = new[] { separatorChar },
HasHeader = hasHeader,
AllowQuoting = allowQuoting,
TrimWhitespace = trimWhitespace,
AllowSparse = allowSparse
};
var loader = new TextLoader(CatalogUtils.GetEnvironment(catalog), options: options);
return(loader.Load(new MultiFileSource(path)));
}
/// <summary>
/// Create a text loader <see cref="TextLoader"/>.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="columns">Array of columns <see cref="TextLoader.Column"/> defining the schema.</param>
/// <param name="separatorChar">The character used as separator between data points in a row. By default the tab character is used as separator.</param>
/// <param name="hasHeader">Whether the file has a header.</param>
/// <param name="allowSparse">Whether the file can contain numerical vectors in sparse format.</param>
/// <param name="allowQuoting">Whether the file can contain column defined by a quoted string.</param>
/// <param name="dataSample">The optional location of a data sample. The sample can be used to infer column names and number of slots in each column.</param>
public static TextLoader CreateTextLoader(this DataOperationsCatalog catalog,
TextLoader.Column[] columns,
char separatorChar = TextLoader.Defaults.Separator,
bool hasHeader = TextLoader.Defaults.HasHeader,
bool allowSparse = TextLoader.Defaults.AllowSparse,
bool allowQuoting = TextLoader.Defaults.AllowQuoting,
IMultiStreamSource dataSample = null)
=> new TextLoader(CatalogUtils.GetEnvironment(catalog), columns, separatorChar, hasHeader, allowSparse, allowQuoting, dataSample);
/// <summary>
/// Load a <see cref="IDataView"/> from a text file using <see cref="TextLoader"/>.
/// Note that <see cref="IDataView"/>'s are lazy, so no actual loading happens here, just schema validation.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="path">Specifies a file or path of files from which to load.</param>
/// <param name="options">Defines the settings of the load operation. No need to specify a Columns field,
/// as columns will be infered by this method.</param>
/// <returns>The data view.</returns>
public static IDataView LoadFromTextFile <TInput>(this DataOperationsCatalog catalog, string path,
TextLoader.Options options)
{
CheckValidPathContents(path);
return(TextLoader.CreateTextLoader <TInput>(CatalogUtils.GetEnvironment(catalog), options)
.Load(new MultiFileSource(path)));
}
/// <summary>
/// Read a data view from a text file using <see cref="TextLoader"/>.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="path">Specifies a file from which to read.</param>
/// <param name="args">Defines the settings of the load operation.</param>
public static IDataView ReadFromTextFile(this DataOperationsCatalog catalog, string path, TextLoader.Arguments args = null)
{
Contracts.CheckNonEmpty(path, nameof(path));
var env = catalog.GetEnvironment();
var source = new MultiFileSource(path);
return(new TextLoader(env, args, source).Read(source));
}
/// <summary>
/// Create a text loader <see cref="TextLoader"/> by inferencing the dataset schema from a data model type.
/// </summary>
/// <typeparam name="TInput">Defines the schema of the data to be loaded. Use public fields or properties
/// decorated with <see cref="LoadColumnAttribute"/> (and possibly other attributes) to specify the column
/// names and their data types in the schema of the loaded data.</typeparam>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="separatorChar">Column separator character. Default is '\t'</param>
/// <param name="hasHeader">Whether the file has a header with feature names. When a <see paramref="dataSample"/> is provided, <see langword="true"/>
/// indicates that the first line in the <see paramref="dataSample"/> will be used for feature names, and that when <see cref="TextLoader.Load(IMultiStreamSource)"/>
/// is called, the first line will be skipped. When there is no <see paramref="dataSample"/> provided, <see langword="true"/> just indicates that the loader should
/// skip the first line when <see cref="TextLoader.Load(IMultiStreamSource)"/> is called, but columns will not have slot names annotations. This is
/// because the output schema is made when the loader is created, and not when <see cref="TextLoader.Load(IMultiStreamSource)"/> is called.</param>
/// <param name="dataSample">The optional location of a data sample. The sample can be used to infer slot name annotations if present.</param>
/// <param name="allowQuoting">Whether the input may include double-quoted values. This parameter is used to distinguish separator characters
/// in an input value from actual separators. When <see langword="true"/>, separators within double quotes are treated as part of the
/// input value. When <see langword="false"/>, all separators, even those whitin quotes, are treated as delimiting a new column.</param>
/// <param name="trimWhitespace">Remove trailing whitespace from lines.</param>
/// <param name="allowSparse">Whether the input may include sparse representations. For example, a row containing
/// "5 2:6 4:3" means that there are 5 columns, and the only non-zero are columns 2 and 4, which have values 6 and 3,
/// respectively. Column indices are zero-based, so columns 2 and 4 represent the 3rd and 5th columns.
/// A column may also have dense values followed by sparse values represented in this fashion. For example,
/// a row containing "1 2 5 2:6 4:3" represents two dense columns with values 1 and 2, followed by 5 sparsely represented
/// columns with values 0, 0, 6, 0, and 3. The indices of the sparse columns start from 0, even though 0 represents the third column.</param>
public static TextLoader CreateTextLoader <TInput>(this DataOperationsCatalog catalog,
char separatorChar = TextLoader.Defaults.Separator,
bool hasHeader = TextLoader.Defaults.HasHeader,
IMultiStreamSource dataSample = null,
bool allowQuoting = TextLoader.Defaults.AllowQuoting,
bool trimWhitespace = TextLoader.Defaults.TrimWhitespace,
bool allowSparse = TextLoader.Defaults.AllowSparse)
=> TextLoader.CreateTextLoader <TInput>(CatalogUtils.GetEnvironment(catalog), hasHeader, separatorChar, allowQuoting,
allowSparse, trimWhitespace, dataSample: dataSample);
/// <summary>
/// Read a data view from an <see cref="IMultiStreamSource"/> on a binary file using <see cref="BinaryLoader"/>.
/// </summary>
/// <param name="catalog">The catalog.</param>
/// <param name="fileSource">The file source to read from. This can be a <see cref="MultiFileSource"/>, for example.</param>
public static IDataView ReadFromBinary(this DataOperationsCatalog catalog, IMultiStreamSource fileSource)
{
Contracts.CheckValue(fileSource, nameof(fileSource));
var env = catalog.GetEnvironment();
var reader = new BinaryLoader(env, new BinaryLoader.Arguments(), fileSource);
return(reader);
}
/// <summary>
/// Create a database loader <see cref="DatabaseLoader"/>.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="columns">Array of columns <see cref="DatabaseLoader.Column"/> defining the schema.</param>
public static DatabaseLoader CreateDatabaseLoader(this DataOperationsCatalog catalog,
params DatabaseLoader.Column[] columns)
{
var options = new DatabaseLoader.Options
{
Columns = columns,
};
return(new DatabaseLoader(CatalogUtils.GetEnvironment(catalog), options));
}
/// <summary>
/// Load a <see cref="IDataView"/> from a text file using <see cref="TextLoader"/>.
/// Note that <see cref="IDataView"/>'s are lazy, so no actual loading happens here, just schema validation.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="path">Specifies a file or path of files from which to load.</param>
/// <param name="options">Defines the settings of the load operation.</param>
/// <example>
/// <format type="text/markdown">
/// <]
/// ]]>
/// </format>
/// </example>
public static IDataView LoadFromTextFile(this DataOperationsCatalog catalog, string path,
TextLoader.Options options = null)
{
CheckValidPathContents(path);
var env = catalog.GetEnvironment();
var source = new MultiFileSource(path);
return(new TextLoader(env, options, dataSample: source).Load(source));
}
/// <summary>Create a database loader <see cref="DatabaseLoader"/>.</summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="columns">Array of columns <see cref="DatabaseLoader.Column"/> defining the schema.</param>
public static DatabaseLoader CreateDatabaseLoader(this DataOperationsCatalog catalog,
params DatabaseLoader.Column[] columns)
{
var options = new DatabaseLoader.Options
{
Columns = columns,
};
return(catalog.CreateDatabaseLoader(options));
}
/// <summary>
/// Read a data view from a binary file using <see cref="BinaryLoader"/>.
/// </summary>
/// <param name="catalog">The catalog.</param>
/// <param name="path">The path to the file to read from.</param>
public static IDataView ReadFromBinary(this DataOperationsCatalog catalog, string path)
{
Contracts.CheckNonEmpty(path, nameof(path));
var env = catalog.GetEnvironment();
var reader = new BinaryLoader(env, new BinaryLoader.Arguments(), path);
return(reader);
}
/// <summary>
/// Load a <see cref="IDataView"/> from a text file using <see cref="TextLoader"/>.
/// Note that <see cref="IDataView"/>'s are lazy, so no actual loading happens here, just schema validation.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="path">Specifies a file from which to load.</param>
/// <param name="options">Defines the settings of the load operation. No need to specify a Columns field,
/// as columns will be infered by this method.</param>
/// <returns>The data view.</returns>
public static IDataView LoadFromTextFile <TInput>(this DataOperationsCatalog catalog, string path,
TextLoader.Options options)
{
Contracts.CheckNonEmpty(path, nameof(path));
if (!File.Exists(path))
{
throw Contracts.ExceptParam(nameof(path), "File does not exist at path: {0}", path);
}
return(TextLoader.CreateTextLoader <TInput>(CatalogUtils.GetEnvironment(catalog), options)
.Load(new MultiFileSource(path)));
}
/// <summary>
/// Save the data view into a binary stream.
/// </summary>
/// <param name="catalog">The catalog.</param>
/// <param name="data">The data view to save.</param>
/// <param name="stream">The stream to write to.</param>
/// <param name="keepHidden">Whether to keep hidden columns in the dataset.</param>
public static void SaveAsBinary(this DataOperationsCatalog catalog, IDataView data, Stream stream,
bool keepHidden = false)
{
Contracts.CheckValue(catalog, nameof(catalog));
Contracts.CheckValue(data, nameof(data));
Contracts.CheckValue(stream, nameof(stream));
var env = catalog.GetEnvironment();
var saver = new BinarySaver(env, new BinarySaver.Arguments());
using (var ch = env.Start("Saving data"))
DataSaverUtils.SaveDataView(ch, saver, data, stream, keepHidden);
}
/// <summary>
/// Load a <see cref="IDataView"/> from a text file using <see cref="TextLoader"/>.
/// Note that <see cref="IDataView"/>'s are lazy, so no actual loading happens here, just schema validation.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="path">Specifies a file from which to load.</param>
/// <param name="options">Defines the settings of the load operation.</param>
/// <example>
/// <format type="text/markdown">
/// <]
/// ]]>
/// </format>
/// </example>
public static IDataView LoadFromTextFile(this DataOperationsCatalog catalog, string path,
TextLoader.Options options = null)
{
Contracts.CheckNonEmpty(path, nameof(path));
if (!File.Exists(path))
{
throw Contracts.ExceptParam(nameof(path), "File does not exist at path: {0}", path);
}
var env = catalog.GetEnvironment();
var source = new MultiFileSource(path);
return(new TextLoader(env, options, dataSample: source).Load(source));
}
/// <summary>
/// Create the ML context.
/// </summary>
/// <param name="seed">Random seed. Set to <c>null</c> for a non-deterministic environment.</param>
/// <param name="conc">Concurrency level. Set to 1 to run single-threaded. Set to 0 to pick automatically.</param>
public MLContext(int?seed = null, int conc = 0)
{
_env = new LocalEnvironment(seed, conc, MakeCompositionContainer);
_env.AddListener(ProcessMessage);
BinaryClassification = new BinaryClassificationCatalog(_env);
MulticlassClassification = new MulticlassClassificationCatalog(_env);
Regression = new RegressionCatalog(_env);
Clustering = new ClusteringCatalog(_env);
Ranking = new RankingCatalog(_env);
Transforms = new TransformsCatalog(_env);
Model = new ModelOperationsCatalog(_env);
Data = new DataOperationsCatalog(_env);
}
/// <summary>
/// Load a <see cref="IDataView"/> from a binary file.
/// Note that <see cref="IDataView"/>'s are lazy, so no actual loading happens here, just schema validation.
/// </summary>
/// <param name="catalog">The catalog.</param>
/// <param name="path">The path to the file to load from.</param>
/// <example>
/// <format type="text/markdown">
/// <]
/// ]]>
/// </format>
/// </example>
public static IDataView LoadFromBinary(this DataOperationsCatalog catalog, string path)
{
Contracts.CheckNonEmpty(path, nameof(path));
if (!File.Exists(path))
{
throw Contracts.ExceptParam(nameof(path), "File does not exist at path: {0}", path);
}
var env = catalog.GetEnvironment();
var loader = new BinaryLoader(env, new BinaryLoader.Arguments(), path);
return(loader);
}
/// <summary>
/// Load a <see cref="IDataView"/> from a text file using <see cref="TextLoader"/>.
/// Note that <see cref="IDataView"/>'s are lazy, so no actual loading happens here, just schema validation.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="path">The path to the file(s).</param>
/// <param name="separatorChar">Column separator character. Default is '\t'</param>
/// <param name="hasHeader">Whether the file has a header. When <see langword="true"/>, the loader will skip the first line when
/// <see cref="TextLoader.Load(IMultiStreamSource)"/> is called.</param>
/// <param name="allowQuoting">Whether the input may include double-quoted values. This parameter is used to distinguish separator characters
/// in an input value from actual separators. When <see langword="true"/>, separators within double quotes are treated as part of the
/// input value. When <see langword="false"/>, all separators, even those whitin quotes, are treated as delimiting a new column.
/// It is also used to distinguish empty values from missing values. When <see langword="true"/>, missing value are denoted by consecutive
/// separators and empty values by \"\". When <see langword="false"/>, empty values are denoted by consecutive separators and missing
/// values by the default missing value for each type documented in <see cref="DataKind"/>.</param>
/// <param name="trimWhitespace">Remove trailing whitespace from lines.</param>
/// <param name="allowSparse">Whether the input may include sparse representations. For example, a row containing
/// "5 2:6 4:3" means that there are 5 columns, and the only non-zero are columns 2 and 4, which have values 6 and 3,
/// respectively. Column indices are zero-based, so columns 2 and 4 represent the 3rd and 5th columns.
/// A column may also have dense values followed by sparse values represented in this fashion. For example,
/// a row containing "1 2 5 2:6 4:3" represents two dense columns with values 1 and 2, followed by 5 sparsely represented
/// columns with values 0, 0, 6, 0, and 3. The indices of the sparse columns start from 0, even though 0 represents the third column.</param>
/// <returns>The data view.</returns>
public static IDataView LoadFromTextFile <TInput>(this DataOperationsCatalog catalog,
string path,
char separatorChar = TextLoader.Defaults.Separator,
bool hasHeader = TextLoader.Defaults.HasHeader,
bool allowQuoting = TextLoader.Defaults.AllowQuoting,
bool trimWhitespace = TextLoader.Defaults.TrimWhitespace,
bool allowSparse = TextLoader.Defaults.AllowSparse)
{
CheckValidPathContents(path);
// REVIEW: it is almost always a mistake to have a 'trainable' text loader here.
// Therefore, we are going to disallow data sample.
return(TextLoader.CreateTextLoader <TInput>(CatalogUtils.GetEnvironment(catalog), hasHeader, separatorChar,
allowQuoting, allowSparse, trimWhitespace).Load(new MultiFileSource(path)));
}
/// <summary>
/// Create the ML context.
/// </summary>
/// <param name="seed">Random seed. Set to <c>null</c> for a non-deterministic environment.</param>
public MLContext(int?seed = null)
{
_env = new LocalEnvironment(seed);
_env.AddListener(ProcessMessage);
BinaryClassification = new BinaryClassificationCatalog(_env);
MulticlassClassification = new MulticlassClassificationCatalog(_env);
Regression = new RegressionCatalog(_env);
Clustering = new ClusteringCatalog(_env);
Ranking = new RankingCatalog(_env);
AnomalyDetection = new AnomalyDetectionCatalog(_env);
Transforms = new TransformsCatalog(_env);
Model = new ModelOperationsCatalog(_env);
Data = new DataOperationsCatalog(_env);
}
/// <summary>
/// Read a data view from a text file using <see cref="TextLoader"/>.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="columns">The columns of the schema.</param>
/// <param name="hasHeader">Whether the file has a header.</param>
/// <param name="separatorChar">The character used as separator between data points in a row. By default the tab character is used as separator.</param>
/// <param name="path">The path to the file.</param>
/// <returns>The data view.</returns>
public static IDataView ReadFromTextFile(this DataOperationsCatalog catalog,
string path,
TextLoader.Column[] columns,
bool hasHeader = TextLoader.DefaultArguments.HasHeader,
char separatorChar = TextLoader.DefaultArguments.Separator)
{
Contracts.CheckNonEmpty(path, nameof(path));
var env = catalog.GetEnvironment();
// REVIEW: it is almost always a mistake to have a 'trainable' text loader here.
// Therefore, we are going to disallow data sample.
var reader = new TextLoader(env, columns, hasHeader, separatorChar, dataSample: null);
return(reader.Read(new MultiFileSource(path)));
}
private protected CrossValidationResult[] CrossValidateTrain(IDataView data, IEstimator <ITransformer> estimator,
int numFolds, string samplingKeyColumn, int?seed = null)
{
Environment.CheckValue(data, nameof(data));
Environment.CheckValue(estimator, nameof(estimator));
Environment.CheckParam(numFolds > 1, nameof(numFolds), "Must be more than 1");
Environment.CheckValueOrNull(samplingKeyColumn);
DataOperationsCatalog.EnsureGroupPreservationColumn(Environment, ref data, ref samplingKeyColumn, seed);
Func <int, CrossValidationResult> foldFunction =
fold =>
{
var trainFilter = new RangeFilter(Environment, new RangeFilter.Options
{
Column = samplingKeyColumn,
Min = (double)fold / numFolds,
Max = (double)(fold + 1) / numFolds,
Complement = true
}, data);
var testFilter = new RangeFilter(Environment, new RangeFilter.Options
{
Column = samplingKeyColumn,
Min = (double)fold / numFolds,
Max = (double)(fold + 1) / numFolds,
Complement = false
}, data);
var model = estimator.Fit(trainFilter);
var scoredTest = model.Transform(testFilter);
return(new CrossValidationResult(model, scoredTest, fold));
};
// Sequential per-fold training.
// REVIEW: we could have a parallel implementation here. We would need to
// spawn off a separate host per fold in that case.
var result = new CrossValidationResult[numFolds];
for (int fold = 0; fold < numFolds; fold++)
{
result[fold] = foldFunction(fold);
}
return(result);
}
/// <summary>
/// Load a <see cref="IDataView"/> from a text file using <see cref="TextLoader"/>.
/// Note that <see cref="IDataView"/>'s are lazy, so no actual loading happens here, just schema validation.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="path">The path to the file.</param>
/// <param name="separatorChar">Column separator character. Default is '\t'</param>
/// <param name="hasHeader">Whether the file has a header with feature names. Note: If a TextLoader is created with hasHeader = true but without a
/// dataSample, then vector columns made by TextLoader will not contain slot name annotations (slots being the elements of the given vector column),
/// because the output schema is made when the TextLoader is made, and not when <see cref="TextLoader.Load(IMultiStreamSource)"/> is called.
/// In addition, the case where dataSample = null and hasHeader = true indicates to the loader that when it is given a file when Load()
/// is called, it needs to skip the first line.</param>
/// <param name="allowQuoting">Whether the input may include quoted values,
/// which can contain separator characters, colons,
/// and distinguish empty values from missing values. When true, consecutive separators
/// denote a missing value and an empty value is denoted by \"\".
/// When false, consecutive separators denote an empty value.</param>
/// <param name="trimWhitespace">Remove trailing whitespace from lines</param>
/// <param name="allowSparse">Whether the input may include sparse representations for example,
/// if one of the row contains "5 2:6 4:3" that's mean there are 5 columns all zero
/// except for 3rd and 5th columns which have values 6 and 3</param>
/// <returns>The data view.</returns>
public static IDataView LoadFromTextFile <TInput>(this DataOperationsCatalog catalog,
string path,
char separatorChar = TextLoader.Defaults.Separator,
bool hasHeader = TextLoader.Defaults.HasHeader,
bool allowQuoting = TextLoader.Defaults.AllowQuoting,
bool trimWhitespace = TextLoader.Defaults.TrimWhitespace,
bool allowSparse = TextLoader.Defaults.AllowSparse)
{
Contracts.CheckNonEmpty(path, nameof(path));
if (!File.Exists(path))
{
throw Contracts.ExceptParam(nameof(path), "File does not exist at path: {0}", path);
}
// REVIEW: it is almost always a mistake to have a 'trainable' text loader here.
// Therefore, we are going to disallow data sample.
return(TextLoader.CreateTextLoader <TInput>(CatalogUtils.GetEnvironment(catalog), hasHeader, separatorChar,
allowQuoting, allowSparse, trimWhitespace).Load(new MultiFileSource(path)));
}
/// <summary>
/// Save the data view as text.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="data">The data view to save.</param>
/// <param name="stream">The stream to write to.</param>
/// <param name="separatorChar">The column separator.</param>
/// <param name="headerRow">Whether to write the header row.</param>
/// <param name="schema">Whether to write the header comment with the schema.</param>
/// <param name="keepHidden">Whether to keep hidden columns in the dataset.</param>
public static void SaveAsText(this DataOperationsCatalog catalog,
IDataView data,
Stream stream,
char separatorChar = TextLoader.DefaultArguments.Separator,
bool headerRow = TextLoader.DefaultArguments.HasHeader,
bool schema = true,
bool keepHidden = false)
{
Contracts.CheckValue(catalog, nameof(catalog));
Contracts.CheckValue(data, nameof(data));
Contracts.CheckValue(stream, nameof(stream));
var env = catalog.GetEnvironment();
var saver = new TextSaver(env, new TextSaver.Arguments {
Separator = separatorChar.ToString(), OutputHeader = headerRow, OutputSchema = schema
});
using (var ch = env.Start("Saving data"))
DataSaverUtils.SaveDataView(ch, saver, data, stream, keepHidden);
}
/// <summary>
/// Create a text loader <see cref="TextLoader"/>.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="columns">Array of columns <see cref="TextLoader.Column"/> defining the schema.</param>
/// <param name="separatorChar">The character used as separator between data points in a row. By default the tab character is used as separator.</param>
/// <param name="hasHeader">Whether the file has a header with feature names. When a <see paramref="dataSample"/> is provided, <see langword="true"/>
/// indicates that the first line in the <see paramref="dataSample"/> will be used for feature names, and that when <see cref="TextLoader.Load(IMultiStreamSource)"/>
/// is called, the first line will be skipped. When there is no <see paramref="dataSample"/> provided, <see langword="true"/> just indicates that the loader should
/// skip the first line when <see cref="TextLoader.Load(IMultiStreamSource)"/> is called, but columns will not have slot names annotations. This is
/// because the output schema is made when the loader is created, and not when <see cref="TextLoader.Load(IMultiStreamSource)"/> is called.</param>
/// <param name="dataSample">The optional location of a data sample. The sample can be used to infer slot name annotations if present, and also the number
/// of slots in a column defined with <see cref="TextLoader.Range"/> with <see langword="null"/> maximum index.
/// If the sample has been saved with ML.NET's <see cref="SaveAsText(DataOperationsCatalog, IDataView, Stream, char, bool, bool, bool, bool)"/>,
/// it will also contain the schema information in the header that the loader can read even if <paramref name="columns"/> is <see langword="null"/>.
/// In order to use the schema defined in the file, all other arguments sould be left with their default values.</param>
/// <param name="allowQuoting">Whether the input may include double-quoted values. This parameter is used to distinguish separator characters
/// in an input value from actual separators. When <see langword="true"/>, separators within double quotes are treated as part of the
/// input value. When <see langword="false"/>, all separators, even those within quotes, are treated as delimiting a new column.</param>
/// <param name="trimWhitespace">Remove trailing whitespace from lines.</param>
/// <param name="allowSparse">Whether the input may include sparse representations. For example, a row containing
/// "5 2:6 4:3" means that there are 5 columns, and the only non-zero are columns 2 and 4, which have values 6 and 3,
/// respectively. Column indices are zero-based, so columns 2 and 4 represent the 3rd and 5th columns.
/// A column may also have dense values followed by sparse values represented in this fashion. For example,
/// a row containing "1 2 5 2:6 4:3" represents two dense columns with values 1 and 2, followed by 5 sparsely represented
/// columns with values 0, 0, 6, 0, and 3. The indices of the sparse columns start from 0, even though 0 represents the third column.</param>
/// <example>
/// <format type="text/markdown">
/// <]
/// ]]>
/// </format>
/// </example>
public static TextLoader CreateTextLoader(this DataOperationsCatalog catalog,
TextLoader.Column[] columns,
char separatorChar = TextLoader.Defaults.Separator,
bool hasHeader = TextLoader.Defaults.HasHeader,
IMultiStreamSource dataSample = null,
bool allowQuoting = TextLoader.Defaults.AllowQuoting,
bool trimWhitespace = TextLoader.Defaults.TrimWhitespace,
bool allowSparse = TextLoader.Defaults.AllowSparse)
{
var options = new TextLoader.Options
{
Columns = columns,
Separators = new[] { separatorChar },
HasHeader = hasHeader,
AllowQuoting = allowQuoting,
TrimWhitespace = trimWhitespace,
AllowSparse = allowSparse,
};
return(new TextLoader(CatalogUtils.GetEnvironment(catalog), options: options, dataSample: dataSample));
}
/// <summary>
/// Create a text loader <see cref="TextLoader"/>.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="options">Defines the settings of the load operation.</param>
/// <param name="dataSample">The optional location of a data sample. The sample can be used to infer slot name annotations if present, and also the number
/// of slots in <see cref="TextLoader.Options.Columns"/> defined with <see cref="TextLoader.Range"/> with <see langword="null"/> maximum index.
/// If the sample has been saved with ML.NET's <see cref="SaveAsText(DataOperationsCatalog, IDataView, Stream, char, bool, bool, bool, bool)"/>,
/// it will also contain the schema information in the header that the loader can read even if <see cref="TextLoader.Options.Columns"/> are not specified.
/// In order to use the schema defined in the file, all other <see cref="TextLoader.Options"/> sould be left with their default values.</param>
public static TextLoader CreateTextLoader(this DataOperationsCatalog catalog,
TextLoader.Options options,
IMultiStreamSource dataSample = null)
=> new TextLoader(CatalogUtils.GetEnvironment(catalog), options, dataSample);
/// <summary>Create a database loader <see cref="DatabaseLoader"/>.</summary> /// <typeparam name="TInput">Defines the schema of the data to be loaded. Use public fields or properties /// decorated with <see cref="LoadColumnAttribute"/> (and possibly other attributes) to specify the column /// names and their data types in the schema of the loaded data.</typeparam> /// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param> public static DatabaseLoader CreateDatabaseLoader <TInput>(this DataOperationsCatalog catalog) => DatabaseLoader.CreateDatabaseLoader <TInput>(CatalogUtils.GetEnvironment(catalog));
/// <summary>
/// Drop rows where a specified predicate returns true.
/// </summary>
/// <typeparam name="TSrc">The class defining which columns to take from the incoming data.</typeparam>
/// <param name="catalog">The data operations catalog.</param>
/// <param name="input">The input data.</param>
/// <param name="filterPredicate">A predicate, that takes an input of type <typeparamref name="TSrc"/> and returns true if the row should be filtered (dropped) and false otherwise.</param>
public static IDataView FilterByCustomPredicate <TSrc>(this DataOperationsCatalog catalog, IDataView input, Func <TSrc, bool> filterPredicate)
where TSrc : class, new()
=> new CustomMappingFilter <TSrc>(catalog.GetEnvironment(), input, filterPredicate);
/// <summary>
/// Create a text loader <see cref="TextLoader"/>.
/// </summary>
/// <param name="catalog">The <see cref="DataOperationsCatalog"/> catalog.</param>
/// <param name="columns">Array of columns <see cref="TextLoader.Column"/> defining the schema.</param>
/// <param name="hasHeader">Whether the file has a header.</param>
/// <param name="separatorChar">The character used as separator between data points in a row. By default the tab character is used as separator.</param>
/// <param name="dataSample">The optional location of a data sample. The sample can be used to infer column names and number of slots in each column.</param>
public static TextLoader CreateTextLoader(this DataOperationsCatalog catalog,
TextLoader.Column[] columns,
bool hasHeader = TextLoader.DefaultArguments.HasHeader,
char separatorChar = TextLoader.DefaultArguments.Separator,
IMultiStreamSource dataSample = null)
=> new TextLoader(CatalogUtils.GetEnvironment(catalog), columns, hasHeader, separatorChar, dataSample);
/// <summary>
/// Drop rows where a specified predicate returns true. This filter allows to maintain a per-cursor state.
/// </summary>
/// <typeparam name="TSrc">The class defining which columns to take from the incoming data.</typeparam>
/// <typeparam name="TState">The type that describes per-cursor state.</typeparam>
/// <param name="catalog">The data operations catalog.</param>
/// <param name="input">The input data.</param>
/// <param name="filterPredicate">A predicate, that takes an input of type <typeparamref name="TSrc"/> and a state object of type
/// <typeparamref name="TState"/>, and returns true if the row should be filtered (dropped) and false otherwise.</param>
/// <param name="stateInitAction">The action to initialize the state object, that is called once before the cursor is initialized.</param>
public static IDataView FilterByStatefulCustomPredicate <TSrc, TState>(this DataOperationsCatalog catalog, IDataView input, Func <TSrc, TState, bool> filterPredicate,
Action <TState> stateInitAction)
where TSrc : class, new()
where TState : class, new()
=> new StatefulCustomMappingFilter <TSrc, TState>(catalog.GetEnvironment(), input, filterPredicate, stateInitAction);
