aboutsummaryrefslogtreecommitdiff
path: root/arm_compute/core/NEON/kernels/NEIm2ColKernel.h
diff options
context:
space:
mode:
authorGiorgio Arena <giorgio.arena@arm.com>2018-07-18 16:13:12 +0100
committerAnthony Barbier <anthony.barbier@arm.com>2018-11-02 16:54:54 +0000
commit0f170396e84836ad8c54d54421e95c61812968be (patch)
treeb8993f251e3c023aca2856b2784e67eb9f11c8a4 /arm_compute/core/NEON/kernels/NEIm2ColKernel.h
parentb6eb35371d222c6b7f61210d97ebd7dd9e197458 (diff)
downloadComputeLibrary-0f170396e84836ad8c54d54421e95c61812968be.tar.gz
COMPMID-1342 Add grouping support to CLIm2ColKernel
Change-Id: I4afb19751520a90fee27fb49b775cd10e92a94f5 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/140476 Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/core/NEON/kernels/NEIm2ColKernel.h')
-rw-r--r--arm_compute/core/NEON/kernels/NEIm2ColKernel.h12
1 files changed, 7 insertions, 5 deletions
diff --git a/arm_compute/core/NEON/kernels/NEIm2ColKernel.h b/arm_compute/core/NEON/kernels/NEIm2ColKernel.h
index 19da7cfd53..37145a38c1 100644
--- a/arm_compute/core/NEON/kernels/NEIm2ColKernel.h
+++ b/arm_compute/core/NEON/kernels/NEIm2ColKernel.h
@@ -83,12 +83,13 @@ public:
* @param[in] kernel_dims The kernel dimensions (width and height).
* @param[in] conv_info Contains padding and stride information described in @ref PadStrideInfo.
* @param[in] has_bias In case biases are provided expands the matrix with 1.
+ * @param[in] dilation (Optional) Dilation, in elements, across x and y. Defaults to (1, 1).
+ * @param[in] num_groups (Optional) Number of groups when performing a grouped convolution
* @param[in] is_fully_connected (Optional) Determines whether this kernel will be called by @ref NEFullyConnectedLayer in order to validate the arguments
* @param[in] is_flatten (Optional) Determines whether this kernel will be called by @ref NEFlattenLayer in order to validate the arguments
- * @param[in] dilation (Optional) Dilation, in elements, across x and y. Defaults to (1, 1).
*/
void configure(const ITensor *input, ITensor *output, const Size2D &kernel_dims, const PadStrideInfo &conv_info,
- bool has_bias, bool is_fully_connected = false, bool is_flatten = false, const Size2D &dilation = Size2D(1U, 1U));
+ bool has_bias, const Size2D &dilation = Size2D(1U, 1U), unsigned int num_groups = 1, bool is_fully_connected = false, bool is_flatten = false);
/** Static function to check if given info will lead to a valid configuration of @ref NEIm2ColKernel
*
* @param[in] input The input tensor to convert. 3 lower dimensions represent a single input [width, height, IFM],
@@ -98,14 +99,15 @@ public:
* @param[in] kernel_dims The kernel dimensions (width and height).
* @param[in] conv_info Contains padding and stride information described in @ref PadStrideInfo.
* @param[in] has_bias In case biases are provided expands the matrix with 1.
- * @param[in] is_fully_connected (Optional) Determines whether this kernel will be called by @ref NEFullyConnectedLayer in order to validate the arguments
- * @param[in] is_flatten (Optional) Determines whether this kernel will be called by @ref NEFlattenLayer in order to validate the arguments
* @param[in] dilation (Optional) Dilation, in elements, across x and y. Defaults to (1, 1).
+ * @param[in] num_groups (Optional) Number of groups when performing a grouped convolution
+ * @param[in] is_fully_connected (Optional)Determines whether this kernel will be called by @ref NEFullyConnectedLayer in order to validate the arguments
+ * @param[in] is_flatten (Optional) Determines whether this kernel will be called by @ref NEFlattenLayer in order to validate the arguments
*
* @return a status
*/
static Status validate(const ITensorInfo *input, const ITensorInfo *output, const Size2D &kernel_dims, const PadStrideInfo &conv_info,
- bool has_bias, bool is_fully_connected = false, bool is_flatten = false, const Size2D &dilation = Size2D(1U, 1U));
+ bool has_bias, const Size2D &dilation = Size2D(1U, 1U), unsigned int num_groups = 1, bool is_fully_connected = false, bool is_flatten = false);
// Inherited methods overridden:
void run(const Window &window, const ThreadInfo &info) override;