EmbeddingModel

An embedding is a numerical representation of a piece of information, for example, text, documents, images, audio, etc. The representation captures the semantic meaning of what is being embedded. This is a meta-model which takes in a embedding_model and a base_model. The embeddings are computed by the embedding_model, and passed as numerical input to the base_model for training/prediction.

Parameters

• embedding_model(embedding_model) → The embedding model to used.
• base_model(base_model) → The base classifier or regressor model.

Example Usage

We can use and deploy the EmbeddingModel as such.

import turboml as tb
embedding = tb.RandomProjectionEmbedding(n_embeddings = 4)
htc_model = tb.HoeffdingTreeClassifier(n_classes=2)
embedding_model = tb.EmbeddingModel(embedding_model = embedding, base_model = htc_model)