Set accumulate flag * * @param[in] accumulate sets whether or not to use accumulation */

Set fast math flag * * @param[in] fast_math Flag to set */

Return the start of the valid region for the given dimension @p d */

take a -128..127 value and stbi__clamp it and convert to 0..255

returns 1 if "a*b*c + add" has no negative terms/factors and doesn't overflow

Hyperbolic tangent. \param arg function argument \return hyperbolic tangent value of \a arg template<typename T> typename enable<expr,T>::type tanh(T

Check if an internal context is valid * * @param[in] ctx Internal context to check * * @return A status code */

Calculate the im2col output shape of a tensor * * @param[in] input Input tensor info * @param[in] kernel_dims The kernel dimensio

Compute the requantization offset and scale * * @deprecated because reequantization using integer offsets creates rounding issues. * Please use @re

The size in bytes of the data type * * @param[in] dt Input data type * * @return The size in bytes of the data type */

! \brief Get error code associated with exception * * \return The error code. */

Gets error description if any * * @return Error description. */

Decompress floating point number. \param arg number to decompress \param exp address to store exponent at \return significant in range [0.5, 1) templa

Multiply by power of two. \param arg number to modify \param exp power of two to multiply with \return \a arg multplied by 2 raised to \a exp template

Flag which specifies whether b should be pre-transposed if supported. * More concretely, the "pre-transpose" is the transposition of the b tensor

Sets GEMMLowp output stage * * @param[in] output_stage Output stage to set */

initialize a callback-based context

initialize a memory-decode context

Target type accessor * * @return Target that the context is associated with */

Check if an internal tensor is valid * * @param[in] tensor Internal tensor to check * * @return A status code */

Flag which specifies if GEMM should accumulate the result in destination or not. * * @return True if GEMM is accumulating the result. *

Returns a read-only (constant) iterator that points to the first element in the dimension array. * * @return an iterator. */

Calculate the gather output shape of a tensor * * @param[in] input_shape Input tensor shape * @param[in] indices_shape Indices tensor shape. Only

Calculate the output pool3d shape of a tensor * * @param[in] src Input tensor info * @param[in] pool3d_info Pooling layer info * * @retur

Calculate the reductionA shape used in GEMMLowp * * @param[in] b Input tensor info * * @return the calculated shape */

Calculate the reductionB shape used in GEMMLowp * * @param[in] a Input tensor info * * @return the calculated shape */

Calculate the stack output shape of a tensor * * @param[in] a Input tensor info * @param[in] axis Axis on which to perform the sta

Calculate the reshaped shape of the weights * * @param[in] weights Weights tensor info * @param[in] has_bias (Optional) Set to true if there i

Calculate the number of output tiles required by Winograd Convolution layer. This utility function can be used by the Winograd input transform * to

Calculate the winograd filter transform shape * * @param[in] input Input tensor info * @param[in] winograd_info Winograd information * *

Calculate the winograd input transform shape * * @param[in] input Input tensor info * @param[in] winograd_info Winograd information * * @

Calculate the winograd output transform shape * * @param[in] input Input tensor info * @param[in] winograd_info Winograd information * *

Cosine function. \param arg function argument \return cosine value of \a arg template<typename T> typename enable<expr,T>::type cos(T arg) { return fu