Element-wise ReLU function

Symbolic element-wise ReLU (Rectified Linear Unit) function for LazyTensor objects.

Usage

# S3 method for LazyTensor
relu(x)

Arguments

x: a LazyTensor, a ComplexLazyTensor, a vector of numeric values, or a scalar value.

Value

An object of class LazyTensor.

Details

If x is a LazyTensor, relu(x) returns a LazyTensor that encodes, symbolically, the element-wise ReLU of x.

The ReLU function (Fukushima 1975) is defined as follows: relu(x) returns 0 if x < 0, x if x >= 0.

References

Fukushima K (1975). “Cognitron: A self-organizing multilayered neural network.” Biological Cybernetics, 20(3), 121--136. ISSN 1432-0770, doi:10.1007/BF00342633 .

Author

Chloe Serre-Combe, Amelie Vernay

Examples

if (FALSE) {
x <- matrix(runif(150 * 3), 150, 3) # arbitrary R matrix, 150 rows, 3 columns
x_i <- LazyTensor(x, index = 'i')   # creating LazyTensor from matrix x, 
                                    # indexed by 'i'
ReLU_x <- relu(x_i)                 # symbolic matrix, 150 rows and 3 columns
}