Module autograd::ops

Modules

dummy_op
random_ops

Traits

Op	Represents a operation node in a computation graph. Tensor wraps trait-object of this.

Functions

acos
acosh
add
add_n
argmax	Takes argmax along specified axis.
asin
asinh
atan
atanh
clip	Limits all elements so as to be within [min, max]
concat	Concat input tensors.
cos
cosh
div
elu	Elementwise exponential linear unit function. (https://arxiv.org/abs/1511.07289)
equals	Returns binary tensor.
exp
expand_dims	Expands dims.
flatten	Returns 1-ranked tensor (vector)
gather	Gather slices.
gradients	Returns gradient tensors wrt variables.
greater	Returns binary tensor.
greater_equal	Returns binary tensor.
identity
lesser	Returns binary tensor.
lesser_equal	Returns binary tensor.
log
log_softmax	Computes log(softmax(x)) along specified axis.
logsumexp	Computes log(sum(exp(x))) along specified axis.
matmul	Matrix multiplication.
mean_squared_error	Just computes 0.5*(a-b)2.
mul
pow
reduce_max	Take max along specified axis.
reduce_mean	Take mean along specified axis.
reduce_min	Take min along specified axis.
reduce_sum	Take sum along specified axis.
relu	Elementwise rectified linear unit function.
reshape	Reshapes input tensor.
rnn_step	Applies recurrent net unit to the input.
sigmoid	Elementwise logistic sigmoid function.
sigmoid_cross_entropy	Computes binary_cross_entropy(sigmoid(y)).
sin
sinh
slice	Slice op.
softmax	Softmax function.
softmax_cross_entropy	Computes categorical_cross_entropy(softmax(y)).
sparse_softmax_cross_entropy	A variant of softmax_cross_entropy.
sqrt
squeeze	Squeezes dims.
sub
swap_axes	Swaps two axes.
tan
tanh
tile	Tiles input tensor along specified axis.