COMPMID-631: Merge branches/gles_compute branch

Last commit:
commit b25c5f68042b0c81bf611d59a1bb8535e1c42497
Author: Xinghang Zhou <xinghang.zhou@arm.com>
Date:   Wed Oct 25 18:48:10 2017 +0800

    Synced validation's tolerances of GCSoftmax from cl side

Change-Id: Ibe72054205c1c8721845d679a31af7ed0a7c5cf6
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/93283
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>

1 files changed, 109 insertions, 0 deletions

diff --git a/arm_compute/core/GLES_COMPUTE/kernels/GCIm2ColKernel.h b/arm_compute/core/GLES_COMPUTE/kernels/GCIm2ColKernel.h
new file mode 100644
index 0000000000..e1b35607ff
--- /dev/null
+++ b/arm_compute/core/GLES_COMPUTE/kernels/GCIm2ColKernel.h
@@ -0,0 +1,109 @@
+/*
+ * Copyright (c) 2017 ARM Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef __ARM_COMPUTE_GCIM2COLKERNEL_H__
+#define __ARM_COMPUTE_GCIM2COLKERNEL_H__
+
+#include "arm_compute/core/GLES_COMPUTE/IGCKernel.h"
+
+namespace arm_compute
+{
+class IGCTensor;
+
+/** Interface for the im2col reshape kernel.
+ *
+ * Rearranges image blocks into columns. It is used to strip out each convolution block to a single column.
+ * It is used to transform a convolution to a plain matrix multiplication.
+ *
+ * For example taking into account the image below and assuming 3x3 image blocks with stride of 1 we have:
+ * @f[
+ * \left( \begin{array}{cccc}
+ * a00 & a01 & a02 & a03 \\
+ * a10 & a11 & a12 & a13 \\
+ * a20 & a21 & a22 & a23 \\
+ * a30 & a31 & a32 & a33 \\
+ * \end{array} \right)
+ * =
+ * \left( \begin{array}{ccccccccc}
+ * a00 & a01 & a02 & a10 & a11 & a12 & a20 & a21 & a22 \\
+ * a01 & a02 & a03 & a11 & a12 & a13 & a21 & a22 & a23 \\
+ * a10 & a11 & a12 & a20 & a21 & a22 & a30 & a31 & a32 \\
+ * a11 & a12 & a13 & a21 & a22 & a23 & a31 & a32 & a33 \\
+ * \end{array} \right)
+ * @f]
+ */
+class GCIm2ColKernel : public IGCKernel
+{
+public:
+    /** Default constructor */
+    GCIm2ColKernel();
+    /** Prevent instances of this class from being copied (As this class contains pointers) */
+    GCIm2ColKernel(const GCIm2ColKernel &) = delete;
+    /** Prevent instances of this class from being copied (As this class contains pointers) */
+    GCIm2ColKernel &operator=(const GCIm2ColKernel &) = delete;
+    /** Allow instances of this class to be moved */
+    GCIm2ColKernel(GCIm2ColKernel &&) = default;
+    /** Allow instances of this class to be moved */
+    GCIm2ColKernel &operator=(GCIm2ColKernel &&) = default;
+    /** Set the input and output of the kernel.
+     *
+     * @param[in]  input       The input tensor to convert. 3 lower dimensions represent a single input [width, height, IFM],
+     *                         while every optional dimension from 4 and above represent a batch of inputs. Data types supported: F32
+     * @param[out] output      The output tensor. First 2 lower dimensions represent a transform of each 3D input,
+     *                         while every dimension above represents a batch. Data types supported: Same as @p input
+     * @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.
+     */
+    void configure(const IGCTensor *input, IGCTensor *output, std::pair<unsigned int, unsigned int> kernel_dims, const PadStrideInfo &conv_info, bool has_bias);
+
+    // Inherited methods overridden:
+    void run(const Window &window) override;
+
+private:
+    /** Run the reshape kernel optimised for the special case (stride is 1, padding is 0 and kernel's low 3 dimensions are same as input)
+     *
+     * @param[in]     window Region on which to execute the kernel. (Must be a valid region of the window returned by window()).
+     * @param[in,out] queue  Command queue on which to enqueue the kernel.
+     */
+    void run_reduced(const Window &window);
+    /** run the generic convolution layer input reshape kernel
+     *
+     * @param[in]     window Region on which to execute the kernel. (Must be a valid region of the window returned by window()).
+     * @param[in,out] queue  Command queue on which to enqueue the kernel.
+     */
+    void run_generic(const Window &window);
+
+    /** Common signature for the kernel to run */
+    using Im2ColFunction = void (GCIm2ColKernel::*)(const Window &);
+
+private:
+    const IGCTensor *_input;
+    IGCTensor       *_output;
+    std::pair<unsigned int, unsigned int> _convolved_dims;
+    unsigned int   _num_elems_processed_per_iteration;
+    Im2ColFunction _run_func;
+};
+}
+
+#endif /*__ARM_COMPUTE_GCIM2COLKERNEL_H__ */