tfjs-layers/src/exports_constraints.ts

/**
 * @license
 * Copyright 2018 Google LLC
 *
 * Use of this source code is governed by an MIT-style
 * license that can be found in the LICENSE file or at
 * https://opensource.org/licenses/MIT.
 * =============================================================================
 */
// tslint:disable-next-line:max-line-length
import {Constraint, MaxNorm, MaxNormArgs, MinMaxNorm, MinMaxNormArgs, NonNeg, UnitNorm, UnitNormArgs} from './constraints';

/**
 * MaxNorm weight constraint.
 *
 * Constrains the weights incident to each hidden unit
 * to have a norm less than or equal to a desired value.
 *
 * References
 *       - [Dropout: A Simple Way to Prevent Neural Networks from Overfitting
 * Srivastava, Hinton, et al.
 * 2014](http://www.cs.toronto.edu/~rsalakhu/papers/srivastava14a.pdf)
 *
 * @doc {heading: 'Constraints',namespace: 'constraints'}
 */
export function maxNorm(args: MaxNormArgs): Constraint {
  return new MaxNorm(args);
}

/**
 * Constrains the weights incident to each hidden unit to have unit norm.
 *
 * @doc {heading: 'Constraints', namespace: 'constraints'}
 */
export function unitNorm(args: UnitNormArgs): Constraint {
  return new UnitNorm(args);
}

/**
 * Constrains the weight to be non-negative.
 *
 * @doc {heading: 'Constraints', namespace: 'constraints'}
 */
export function nonNeg(): Constraint {
  return new NonNeg();
}

/** @doc {heading: 'Constraints', namespace: 'constraints'} */
export function minMaxNorm(config: MinMaxNormArgs): Constraint {
  return new MinMaxNorm(config);
}