| File: EntryPoints\PredictorModel.cs | Web Access |
| Project: src\src\Microsoft.ML.Core\Microsoft.ML.Core.csproj (Microsoft.ML.Core) |
// Licensed to the .NET Foundation under one or more agreements. // The .NET Foundation licenses this file to you under the MIT license. // See the LICENSE file in the project root for more information. using System.IO; using Microsoft.ML.Data; using Microsoft.ML.Runtime; namespace Microsoft.ML.EntryPoints; /// <summary> /// Base type for standard predictor model port type. /// </summary> [BestFriend] internal abstract class PredictorModel { [BestFriend] private protected PredictorModel() { } /// <summary> /// Save the model to the given stream. /// </summary> [BestFriend] internal abstract void Save(IHostEnvironment env, Stream stream); /// <summary> /// Extract only the transform portion of the predictor model. /// </summary> [BestFriend] internal abstract TransformModel TransformModel { get; } /// <summary> /// Extract the predictor object out of the predictor model. /// </summary> [BestFriend] internal abstract IPredictor Predictor { get; } /// <summary> /// Apply the predictor model to the transform model and return the resulting predictor model. /// </summary> [BestFriend] internal abstract PredictorModel Apply(IHostEnvironment env, TransformModel transformModel); /// <summary> /// For a given input data, return role mapped data and the predictor object. /// The scoring entry point will hopefully know how to construct a scorer out of them. /// </summary> [BestFriend] internal abstract void PrepareData(IHostEnvironment env, IDataView input, out RoleMappedData roleMappedData, out IPredictor predictor); /// <summary> /// Returns a string array containing the label names of the label column type predictor was trained on. /// If the training label is a key with text key value annotation, it should return this annotation. The order of the labels should be consistent /// with the key values. Otherwise, it returns null. /// </summary> /// <param name="env"/> /// <param name="labelType">The column type of the label the predictor was trained on.</param> [BestFriend] internal abstract string[] GetLabelInfo(IHostEnvironment env, out DataViewType labelType); /// <summary> /// Returns the <see cref="RoleMappedSchema"/> that was used in training. /// </summary> [BestFriend] internal abstract RoleMappedSchema GetTrainingSchema(IHostEnvironment env); }