API Reference tangram v0.7.0

This is the main module in the tangram package.

This describes the contribution of a feature from a bag of words cosine similarity feature group.

This describes the contribution of a feature from a bag of words feature group.

Tangram.predict outputs {:binary_classification, BinaryClassificationPredictOutput.t()} when the model's task is binary classification.

This is a description of the feature contributions for the prediction if the task is regression or binary classification, or for a single class if the task is multiclass classification.

This describes the contribution of a feature from an identity feature group.

These are the options passed when loading a model.

This is the type of the argument to Tangram.log_prediction and Tangram.enqueue_log_prediction which specifies the details of the prediction to log.

This is the type of the argument to Tangram.log_true_value and Tangram.enqueue_log_true_value which specifies the details of the true value to log.

Use this struct to load a model, make predictions, and log events to the app.

Tangram.predict outputs {:multiclass_classification, MulticlassClassificationPredictOutput.t()} when the model's task is multiclass classification.

This describes the contribution of a feature from a normalized feature group.

This describes the contribution of a feature from a one hot encoded feature group.

These are the options passed to Tangram.predict.

Tangram.predict outputs {:regression, RegressionPredictOutput.t()} when the model's task is regression.

This describes the contribution of a feature from a word embedding feature group.