All Downloads are FREE. Search and download functionalities are using the official Maven repository.

scripts.nn.layers.relu.dml Maven / Gradle / Ivy

There is a newer version: 1.2.0
Show newest version
#-------------------------------------------------------------
#
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements.  See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership.  The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License.  You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied.  See the License for the
# specific language governing permissions and limitations
# under the License.
#
#-------------------------------------------------------------

/*
 * Rectified Linear Unit (ReLU) nonlinearity layer.
 */

forward = function(matrix[double] X)
    return (matrix[double] out) {
  /*
   * Computes the forward pass for a ReLU nonlinearity layer.
   *
   * Performs an element-wise evaluation of `f(input) = max(0, input)`.
   *
   * Inputs:
   *  - X: Inputs, of shape (any, any).
   *
   * Outputs:
   *  - out: Outputs, of same shape as `X`.
   */
  out = max(0, X)
}

backward = function(matrix[double] dout, matrix[double] X)
    return (matrix[double] dX) {
  /*
   * Computes the backward pass for a ReLU nonlinearity layer.
   *
   * Essentially performs a pass-through of the upstream gradient
   * for cells > 0.
   *
   * Inputs:
   *  - dout: Gradient wrt `out` from upstream, of same shape as `X`.
   *  - X: Previous input data matrix, of shape (any, any).
   *
   * Outputs:
   *  - dX: Gradient wrt `X`, of same shape as `X`.
   */
   dX = (X > 0) * dout
}





© 2015 - 2024 Weber Informatics LLC | Privacy Policy