Dela via


IDataView Interface

Definition

The input and output of Query Operators (Transforms). This is the fundamental data pipeline type, comparable to IEnumerable<T> for LINQ.

public interface IDataView
type IDataView = interface
Public Interface IDataView
Derived

Examples

using System;
using System.Collections.Generic;
using System.Linq;
using Microsoft.ML;
using Microsoft.ML.Data;

namespace Samples.Dynamic
{
    /// <summary>
    /// The <see cref="IDataView"/> interface is the central concept of "data" in
    /// ML.NET. While many conveniences exist to create pre-baked implementations,
    /// it is also useful to know how to create one completely from scratch. We also
    /// take this opportunity to illustrate and motivate the basic principles of how
    /// the IDataView system is architected, since people interested in
    /// implementing <see cref="IDataView"/> need at least some knowledge of those
    /// principles.
    /// </summary>
    public static class SimpleDataViewImplementation
    {
        public static void Example()
        {
            // First we create an array of these objects, which we "present" as this
            // IDataView implementation so that it can be used in a simple ML.NET
            // pipeline.
            var inputArray = new[]
            {
                new InputObject(false, "Hello my friend."),
                new InputObject(true, "Stay awhile and listen."),
                new InputObject(true, "Masterfully done hero!")
            };
            var dataView = new InputObjectDataView(inputArray);

            // So, this is a very simple pipeline: a transformer that tokenizes
            // Text, does nothing with the Label column at all.
            var mlContext = new MLContext();
            var transformedDataView = mlContext.Transforms.Text.TokenizeIntoWords(
                "TokenizedText", "Text").Fit(dataView).Transform(dataView);

            var textColumn = transformedDataView.Schema["Text"];
            var tokensColumn = transformedDataView.Schema["TokenizedText"];

            using (var cursor = transformedDataView.GetRowCursor(
                new[] { textColumn, tokensColumn }))

            {
                // Note that it is best to get the getters and values *before*
                // iteration, so as to facilitate buffer sharing (if applicable),
                // and column-type validation once, rather than many times.
                ReadOnlyMemory<char> textValue = default;
                VBuffer<ReadOnlyMemory<char>> tokensValue = default;

                var textGetter = cursor
                    .GetGetter<ReadOnlyMemory<char>>(textColumn);

                var tokensGetter = cursor
                    .GetGetter<VBuffer<ReadOnlyMemory<char>>>(tokensColumn);

                while (cursor.MoveNext())
                {
                    textGetter(ref textValue);
                    tokensGetter(ref tokensValue);

                    Console.WriteLine(
                        $"{textValue} => " +
                        $"{string.Join(", ", tokensValue.DenseValues())}");

                }

                // The output to console is this:

                // Hello my friend. => Hello, my, friend.
                // Stay awhile and listen. => Stay, awhile, and, listen.
                // Masterfully done hero! => Masterfully, done, hero!

                // Note that it may be interesting to set a breakpoint on the
                // Console.WriteLine, and explore what is going on with the cursor,
                // and the buffers. In particular, on the third iteration, while
                // `tokensValue` is logically presented as a three element array,
                // internally you will see that the arrays internal to that
                // structure have (at least) four items, specifically:
                // `Masterfully`, `done`, `hero!`, `listen.`. In this way we see a
                // simple example of the details of how buffer sharing from one
                // iteration to the next actually works.
            }
        }

        private sealed class InputObject
        {
            public bool Label { get; }
            public string Text { get; }

            public InputObject(bool label, string text)
            {
                Label = label;
                Text = text;
            }
        }

        /// <summary>
        /// This is an implementation of <see cref="IDataView"/> that wraps an
        /// <see cref="IEnumerable{T}"/> of the above <see cref="InputObject"/>.
        /// Note that normally under these circumstances, the first recommendation
        /// would be to use a convenience like 
        /// <see cref="DataOperationsCatalog
        /// .LoadFromEnumerable{TRow}(IEnumerable{TRow}, SchemaDefinition)"/>
        /// or something like that, rather than implementing <see cref="IDataView"/>
        /// outright. However, sometimes when code generation is impossible on some
        /// situations, like Unity or other similar platforms, implementing
        /// something even closely resembling this may become necessary.
        ///
        /// This implementation of <see cref="IDataView"/>, being didactic, is much
        /// simpler than practically anything one would find in the ML.NET codebase.
        /// In this case we have a completely fixed schema (the two fields of
        /// <see cref="InputObject"/>), with fixed types.
        ///
        /// For <see cref="Schema"/>, note that we keep a very simple schema based
        /// off the members of the object. You may in fact note that it is possible
        /// in this specific case, this implementation of <see cref="IDatView"/>
        /// could share the same <see cref="DataViewSchema"/> object across all
        /// instances of this object, but since this is almost never the case, I do
        /// not take advantage of that.
        ///
        /// We have chosen to wrap an <see cref="IEnumerable{T}"/>, so in fact only
        /// a very simple implementation is possible. Specifically: we cannot
        /// meaningfully shuffle (so <see cref="CanShuffle"/> is
        /// <see langword="false"/>, and even if a <see cref="Random"/>
        /// parameter were passed to
        /// <see cref="GetRowCursor(IEnumerable{DataViewSchema.Column}, Random)"/>,
        /// we could not make use of it), we do not know the count of the item right
        /// away without counting (so, it is most correct for
        /// <see cref="GetRowCount"/> to return <see langword="null"/>, even after
        /// we might hypothetically know after the first pass, given the
        /// immutability principle of <see cref="IDatView"/>), and the
        /// <see cref="GetRowCursorSet(
        /// IEnumerable{DataViewSchema.Column}, int, Random)"/> method returns a
        /// single item.
        ///
        /// The <see cref="DataViewRowCursor"/> derived class has more documentation
        /// specific to its behavior.
        ///
        /// Note that this implementation, as well as the nested
        /// <see cref="DataViewRowCursor"/> derived class, does almost no validation
        /// of parameters or guard against misuse than we would like from, say,
        /// implementations of the same classes within the ML.NET codebase.
        /// </summary>
        private sealed class InputObjectDataView : IDataView
        {
            private readonly IEnumerable<InputObject> _data;
            public IEnumerable<InputObject> Data
            {
                get
                {
                    return _data;
                }
            }
            public DataViewSchema Schema { get; }
            public bool CanShuffle => false;

            public InputObjectDataView(IEnumerable<InputObject> data)
            {
                _data = data;

                var builder = new DataViewSchema.Builder();
                builder.AddColumn("Label", BooleanDataViewType.Instance);
                builder.AddColumn("Text", TextDataViewType.Instance);
                Schema = builder.ToSchema();
            }

            public long? GetRowCount() => null;

            public DataViewRowCursor GetRowCursor(
                IEnumerable<DataViewSchema.Column> columnsNeeded,
                Random rand = null)

                => new Cursor(this, columnsNeeded.Any(c => c.Index == 0),
                    columnsNeeded.Any(c => c.Index == 1));

            public DataViewRowCursor[] GetRowCursorSet(
                IEnumerable<DataViewSchema.Column> columnsNeeded, int n,
                Random rand = null)

                => new[] { GetRowCursor(columnsNeeded, rand) };

            /// <summary>
            /// Having this be a private sealed nested class follows the typical
            /// pattern: in most <see cref="IDataView"/> implementations, the cursor
            /// instance is almost always that. The only "common" exceptions to this
            /// tendency are those implementations that are such thin wrappings of
            /// existing <see cref="IDataView"/> without even bothering to change
            /// the schema.
            ///
            /// On the subject of schema, note that there is an expectation that
            /// the <see cref="Schema"/> object is reference equal to the
            /// <see cref="IDataView.Schema"/> object that created this cursor, as
            /// we see here.
            ///
            /// Note that <see cref="Batch"/> returns <c>0</c>. As described in the
            /// documentation of that property, that is meant to facilitate the
            /// reconciliation of the partitioning of the data in the case where
            /// multiple cursors are returned from
            /// <see cref="GetRowCursorSet(
            /// IEnumerable{DataViewSchema.Column}, int, Random)"/>, 
            /// but since only one is ever returned from the implementation, this
            /// behavior is appropriate.
            ///
            /// Similarly, since it is impossible to have a shuffled cursor or a
            /// cursor set, it is sufficient for the <see cref="GetIdGetter"/>
            /// implementation to return a simple ID based on the position. If,
            /// however, this had been something built on, hypothetically, an
            /// <see cref="IList{T}"/> or some other such structure, and shuffling
            /// and partitioning was available, an ID based on the index of whatever
            /// item was being returned would be appropriate.
            ///
            /// Note the usage of the <see langword="ref"/> parameters on the
            /// <see cref="ValueGetter{TValue}"/> implementations. This is most
            /// valuable in the case of buffer sharing for <see cref="VBuffer{T}"/>,
            /// but we still of course have to deal with it here.
            ///
            /// Note also that we spend a considerable amount of effort to not make
            /// the <see cref="GetGetter{TValue}(DataViewSchema.Column)"/> and
            /// <see cref="IsColumnActive(DataViewSchema.Column)"/> methods
            /// correctly reflect what was asked for from the
            /// <see cref="GetRowCursor(
            /// IEnumerable{DataViewSchema.Column}, Random)"/> method that was used
            /// to create this method. In this particular case, the point is
            /// somewhat moot: this mechanism exists to enable lazy evaluation,
            /// but since this cursor is implemented to wrap an
            /// <see cref="IEnumerator{T}"/> which has no concept of lazy
            /// evaluation, there is no real practical benefit to doing this.
            /// However, it is best of course to illustrate the general principle
            /// for the sake of the example.
            ///
            /// Even in this simple form, we see the reason why
            /// <see cref="GetGetter{TValue}(DataViewSchema.Column)"/> is
            /// beneficial: the <see cref="ValueGetter{TValue}"/> implementations
            /// themselves are simple to the point where their operation is dwarfed
            /// by the simple acts of casting and validation checking one sees in
            /// <see cref="GetGetter{TValue}(DataViewSchema.Column)"/>. In this way
            /// we only pay the cost of validation and casting once, not every time
            /// we get a value.
            /// </summary>
            private sealed class Cursor : DataViewRowCursor
            {
                private bool _disposed;
                private long _position;
                private readonly IEnumerator<InputObject> _enumerator;
                private readonly Delegate[] _getters;

                public override long Position => _position;
                public override long Batch => 0;
                public override DataViewSchema Schema { get; }

                public Cursor(InputObjectDataView parent, bool wantsLabel,
                    bool wantsText)

                {
                    Schema = parent.Schema;
                    _position = -1;
                    _enumerator = parent.Data.GetEnumerator();
                    _getters = new Delegate[]
                    {
                        wantsLabel ?
                            (ValueGetter<bool>)LabelGetterImplementation : null,

                        wantsText ?
                            (ValueGetter<ReadOnlyMemory<char>>)
                            TextGetterImplementation : null

                    };
                }

                protected override void Dispose(bool disposing)
                {
                    if (_disposed)
                        return;
                    if (disposing)
                    {
                        _enumerator.Dispose();
                        _position = -1;
                    }
                    _disposed = true;
                    base.Dispose(disposing);
                }

                private void LabelGetterImplementation(ref bool value)
                    => value = _enumerator.Current.Label;

                private void TextGetterImplementation(
                    ref ReadOnlyMemory<char> value)

                    => value = _enumerator.Current.Text.AsMemory();

                private void IdGetterImplementation(ref DataViewRowId id)
                    => id = new DataViewRowId((ulong)_position, 0);

                public override ValueGetter<TValue> GetGetter<TValue>(
                    DataViewSchema.Column column)

                {
                    if (!IsColumnActive(column))
                        throw new ArgumentOutOfRangeException(nameof(column));
                    return (ValueGetter<TValue>)_getters[column.Index];
                }

                public override ValueGetter<DataViewRowId> GetIdGetter()
                    => IdGetterImplementation;

                public override bool IsColumnActive(DataViewSchema.Column column)
                    => _getters[column.Index] != null;

                public override bool MoveNext()
                {
                    if (_disposed)
                        return false;
                    if (_enumerator.MoveNext())
                    {
                        _position++;
                        return true;
                    }
                    Dispose();
                    return false;
                }
            }
        }
    }
}

Properties

CanShuffle

Whether this IDataView supports shuffling of rows, to any degree.

Schema

Gets an instance of Schema.

Methods

GetRowCount()

Returns the number of rows if known. Returning null means that the row count is unknown but it might return a non-null value on a subsequent call. This indicates, that the transform does not YET know the number of rows, but may in the future. Its implementation's computation complexity should be O(1).

Most implementation will return the same answer every time. Some, like a cache, might return null until the cache is fully populated.

GetRowCursor(IEnumerable<DataViewSchema.Column>, Random)

Get a row cursor. The columnsNeeded indicate the active columns that are needed to iterate over. If set to an empty IEnumerable no column is requested. The schema of the returned cursor will be the same as the schema of the IDataView, but getting a getter for inactive columns will throw.

GetRowCursorSet(IEnumerable<DataViewSchema.Column>, Int32, Random)

This constructs a set of parallel batch cursors. The value n is a recommended limit on cardinality. If n is non-positive, this indicates that the caller has no recommendation, and the implementation should have some default behavior to cover this case. Note that this is strictly a recommendation: it is entirely possible that an implementation can return a different number of cursors.

The cursors should return the same data as returned through GetRowCursor(IEnumerable<DataViewSchema.Column>, Random), except partitioned: no two cursors should return the "same" row as would have been returned through the regular serial cursor, but all rows should be returned by exactly one of the cursors returned from this cursor. The cursors can have their values reconciled downstream through the use of the Batch property.

The typical usage pattern is that a set of cursors is requested, each of them is then given to a set of working threads that consume from them independently while, ultimately, the results are finally collated in the end by exploiting the ordering of the Batch property described above. More typical scenarios will be content with pulling from the single serial cursor of GetRowCursor(IEnumerable<DataViewSchema.Column>, Random).

Extension Methods

GetColumn<T>(IDataView, DataViewSchema+Column)

Extract all values of one column of the data view in a form of an IEnumerable<T>.

GetColumn<T>(IDataView, String)

Extract all values of one column of the data view in a form of an IEnumerable<T>.

Preview(IDataView, Int32)

Extract a 'head' of the data view in a view that is convenient to debug.

Applies to