Incremental Learning\Training

[MaxAkbar]

Is there a possibility of adding incremental training a model?

[Zruty0]

It is definitely possible using the low-level API: some linear models can be 'incrementally trained' using 'initial predictor' to initialize themselves. In #653 there is a code sample (look for TrainWithInitialPredictor test).

We are working on making a high-level API that would support such a scenario: see #584 for the list of things we want to enable.

For instance, this is how currently (in my local branch) I perform incremental training:

                // Pipeline.
                var pipeline = new MyTextLoader(env, MakeSentimentTextLoaderArgs())
                    .Append(new MyTextTransform(env, MakeSentimentTextTransformArgs()));

                // Train the pipeline, prepare train set.
                var reader = pipeline.Fit(new MultiFileSource(dataPath));
                var trainData = reader.Read(new MultiFileSource(dataPath));


                // Train the first predictor.
                var trainer = new MySdca(env, new LinearClassificationTrainer.Arguments
                {
                    NumThreads = 1
                }, "Features", "Label");
                var firstPredictor = trainer.Fit(trainData);

                // Train the second predictor on the same data.
                var secondTrainer = new MyAveragedPerceptron(env, new AveragedPerceptronTrainer.Arguments(), "Features", "Label");
                var finalPredictor = secondTrainer.Train(trainData, firstPredictor.InnerModel);

[MaxAkbar]

Thank you this is very helpful 👍 will take a look.