/* @flow */

/**
 * This is a single-layer [Perceptron Classifier](http://en.wikipedia.org/wiki/Perceptron) that takes
 * arrays of numbers and predicts whether they should be classified
 * as either 0 or 1 (negative or positive examples).
 * @class
 * @example
 * // Create the model
 * var p = new PerceptronModel();
 * // Train the model with input with a diagonal boundary.
 * for (var i = 0; i < 5; i++) {
 *     p.train([1, 1], 1);
 *     p.train([0, 1], 0);
 *     p.train([1, 0], 0);
 *     p.train([0, 0], 0);
 * }
 * p.predict([0, 0]); // 0
 * p.predict([0, 1]); // 0
 * p.predict([1, 0]); // 0
 * p.predict([1, 1]); // 1
 */
function PerceptronModel() {
    // The weights, or coefficients of the model;
    // weights are only populated when training with data.
    this.weights = [];
    // The bias term, or intercept; it is also a weight but
    // it's stored separately for convenience as it is always
    // multiplied by one.
    this.bias = 0;
}

/**
 * **Predict**: Use an array of features with the weight array and bias
 * to predict whether an example is labeled 0 or 1.
 *
 * @param {Array<number>} features an array of features as numbers
 * @returns {number} 1 if the score is over 0, otherwise 0
 */
PerceptronModel.prototype.predict = function(features) {
    // Only predict if previously trained
    // on the same size feature array(s).
    if (features.length !== this.weights.length) {
        return null;
    }

    // Calculate the sum of features times weights,
    // with the bias added (implicitly times one).
    let score = 0;
    for (let i = 0; i < this.weights.length; i++) {
        score += this.weights[i] * features[i];
    }
    score += this.bias;

    // Classify as 1 if the score is over 0, otherwise 0.
    if (score > 0) {
        return 1;
    } else {
        return 0;
    }
};

/**
 * **Train** the classifier with a new example, which is
 * a numeric array of features and a 0 or 1 label.
 *
 * @param {Array<number>} features an array of features as numbers
 * @param {number} label either 0 or 1
 * @returns {PerceptronModel} this
 */
PerceptronModel.prototype.train = function(features, label) {
    // Require that only labels of 0 or 1 are considered.
    if (label !== 0 && label !== 1) {
        return null;
    }
    // The length of the feature array determines
    // the length of the weight array.
    // The perceptron will continue learning as long as
    // it keeps seeing feature arrays of the same length.
    // When it sees a new data shape, it initializes.
    if (features.length !== this.weights.length) {
        this.weights = features;
        this.bias = 1;
    }
    // Make a prediction based on current weights.
    const prediction = this.predict(features);
    // Update the weights if the prediction is wrong.
    if (prediction !== label) {
        const gradient = label - prediction;
        for (let i = 0; i < this.weights.length; i++) {
            this.weights[i] += gradient * features[i];
        }
        this.bias += gradient;
    }
    return this;
};

export default PerceptronModel;