Module dfdx::tensor_ops

Implementations of all operations for tensors, including activations, binary operations, and other methods.

Functions

abs

The absolute value (abs) computes |x|

add

Add two Tensors of the same shape together: &lhs + rhs

binary_map

Applies a binary function f, it’s partial wrt. x dfdx, and its partial wrt. y dfdy to a pair of Tensors lhs and `rhs.

broadcast_inner_sub

Broadcasts the last dimension of rhs to make it the same size of lhs.

broadcast_outer_add

Broadcast the first dimension of Rhs M times, so its the same size as Lhs.

clamp

Clamps all values in t to between min and max

cloned_map

Similar to map(), but doesn’t take ownership of the Tensor t.

cos

The cos function computes cos(x)

div

Divides two Tensors of the same shape: &lhs / rhs.

exp

The exponential function (exp) computes e ^ x

ln

The Natural Logarithm (ln) computes ln(x)

log_softmax

Numerically stable computation of log(softmax(t)). Does t - logsumexp(t) under the hood.

logsumexp

Computes the LogSumExp function. Equivalent to log(sum(exp(data))) or data.exp().sum(-1).log().

map

Applies a function f to every element of the Tensor. The derivative df must also be provided.

matmul

Matrix multiplication.

mean

Sums all the values in self and divides by number of values.

mul

Multiplies two Tensors of the same shape together: &lhs * rhs.

nans_to

Replaces any nans in t with value.

negate

Negates all values in t.

relu

Rectified Linear Unit (ReLU) computes max(0, x).

sigmoid

Sigmoid computes 1 / (1 + exp(-x)).

sin

The sine function computes sin(x)

softmax

Computes the softmax function. Equivalent to t.log_softmax().exp() or exp(log_softmax(t)) or exp(t) / sum(exp(t))

sqrt

Square root computes x ^ 0.5 or √x.

square

Square computes x * x.

sub

Subtracts two Tensors of the same shape from each other: &lhs - rhs

sum_last_dim

Calls [Device::sum_last_dim()] on the underlying array. Result Tensor has smaller number of dimensions.

tanh

Hyperbolic Tangent (Tanh) computes tanh(x).

value_mask

Sets t to value anywhere mask equals value

vecmat_mul

vector * matrix multiplication.