COMPMID-3694 COMPMID-3695 COMPMID-3458: Softmax Axis

* Properly support "axis" in CL and NEON (and GC) SoftmaxLayer and
LogSoftmaxLayer in accord with mainstream frameworks. Axis now defines
the dimension on which softmax is performed, and supports the range
[-rank, rank)

* Extend validation tests to include valid and invalid axes

* Remove unnecessary LogSoftmaxLayer fixture, as it is only a
specialisation of the SoftmaxLayer fixture

* Change the validation fill value range from [-1000, 1000] to [-10,
10], as the former often results in sparse outputs with a single one and
zeros elsewhere

Change-Id: I8a0040453182b04ed88260de3ba434e98258d863
Signed-off-by: Manuel Bottini <manuel.bottini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3830
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>

1 files changed, 48 insertions, 41 deletions

diff --git a/tests/validation/reference/SoftmaxLayer.cpp b/tests/validation/reference/SoftmaxLayer.cpp
index 00206766f8..3fbac32a9b 100644
--- a/tests/validation/reference/SoftmaxLayer.cpp
+++ b/tests/validation/reference/SoftmaxLayer.cpp
@@ -25,6 +25,7 @@
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Types.h"
+#include "utils/TypePrinter.h"
 
 namespace arm_compute
 {
@@ -35,39 +36,43 @@ namespace validation
 namespace reference
 {
 template <typename T, typename std::enable_if<is_floating_point<T>::value, int>::type>
-SimpleTensor<T> softmax_layer_generic(const SimpleTensor<T> &src, float beta, int32_t reduce_end_axis, bool is_log)
+SimpleTensor<T> softmax_layer_generic(const SimpleTensor<T> &src, float beta, int32_t axis, bool is_log)
 {
     // Create reference
     SimpleTensor<T> dst{ src.shape(), src.data_type(), 1 };
 
-    // Convert reduce-before axis (inclusive) to first n axes to reduce
-    const size_t first_n_reduce_axes = dim_index_2_num_dims(reduce_end_axis, static_cast<int32_t>(src.shape().num_dimensions()));
+    const int32_t n_dims = static_cast<int32_t>(src.shape().num_dimensions());
+    ARM_COMPUTE_ERROR_ON(axis < -n_dims || axis >= n_dims);
 
-    // Compute reference. Lower dims are the collapsing of the first axis
-    // dimensions (i.e., the flattened dimension of each batch). The upper dims are
-    // instead the batches we want to normalize
+    const unsigned int actual_axis = static_cast<unsigned int>(wrap_around(axis, n_dims));
+    Window             window;
+    window.use_tensor_dimensions(src.shape());
+    const unsigned int axis_dimension = src.shape()[actual_axis];
+    window.set(actual_axis, Window::Dimension(0, 1, 1));
 
-    const int lower_dims = src.shape().total_size_lower(first_n_reduce_axes);
-
-    const int upper_dims = src.shape().total_size_upper(first_n_reduce_axes);
-
-#if defined(_OPENMP)
-    #pragma omp parallel for
-#endif /* _OPENMP */
-    for(int r = 0; r < upper_dims; ++r)
+    execute_window_loop(window, [&](const Coordinates & id)
     {
-        const T *src_row_ptr = src.data() + r * lower_dims;
-        T       *dst_row_ptr = dst.data() + r * lower_dims;
-
-        // Find max
-        const T max = *std::max_element(src_row_ptr, src_row_ptr + lower_dims);
+        // Find max along axis
+        Coordinates offset(id);
+        offset.set(actual_axis, 0);
+        T max = *reinterpret_cast<const T *>(src(offset));
+        for(unsigned int axis_id = 1; axis_id < axis_dimension; ++axis_id)
+        {
+            offset.set(actual_axis, axis_id);
+            const T val = *reinterpret_cast<const T *>(src(offset));
+            if(val > max)
+            {
+                max = val;
+            }
+        }
 
         // Regularize
         T sum(0.f);
-        std::transform(src_row_ptr, src_row_ptr + lower_dims, dst_row_ptr, [&sum, max, beta, is_log](T val)
+        for(unsigned int axis_id = 0; axis_id < axis_dimension; ++axis_id)
         {
-            T res{ (val - max) *beta };
-
+            offset.set(actual_axis, axis_id);
+            const T val = *reinterpret_cast<const T *>(src(offset));
+            T       res{ (val - max) *beta };
             if(is_log)
             {
                 sum += std::exp(res);
@@ -77,50 +82,52 @@ SimpleTensor<T> softmax_layer_generic(const SimpleTensor<T> &src, float beta, in
                 res = std::exp(res);
                 sum += res;
             }
-            return res;
-        });
+            *reinterpret_cast<T *>(dst(offset)) = res;
+        }
 
         // Normalize
-        std::transform(dst_row_ptr, dst_row_ptr + lower_dims, dst_row_ptr, [sum, is_log](T val)
+        for(unsigned int axis_id = 0; axis_id < axis_dimension; ++axis_id)
         {
+            offset.set(actual_axis, axis_id);
+            const T val = *reinterpret_cast<const T *>(dst(offset));
             if(is_log)
             {
-                return val - static_cast<T>(std::log(sum));
+                *reinterpret_cast<T *>(dst(offset)) = val - static_cast<T>(std::log(sum));
             }
             else
             {
-                return val / sum;
+                *reinterpret_cast<T *>(dst(offset)) = val / sum;
             }
-        });
-    }
-
+        }
+    });
     return dst;
 }
 
-template SimpleTensor<float> softmax_layer_generic(const SimpleTensor<float> &src, float beta, int32_t reduce_end_axis, bool is_log);
-template SimpleTensor<half> softmax_layer_generic(const SimpleTensor<half> &src, float beta, int32_t reduce_end_axis, bool is_log);
+template SimpleTensor<float> softmax_layer_generic(const SimpleTensor<float> &src, float beta, int32_t axis, bool is_log);
+template SimpleTensor<half> softmax_layer_generic(const SimpleTensor<half> &src, float beta, int32_t axis, bool is_log);
 
 template <typename T, typename std::enable_if<is_floating_point<T>::value, int>::type>
-SimpleTensor<T> softmax_layer(const SimpleTensor<T> &src, float beta, int32_t reduce_end_axis)
+SimpleTensor<T> softmax_layer(const SimpleTensor<T> &src, float beta, int32_t axis, bool is_log)
 {
-    return softmax_layer_generic<T>(src, beta, reduce_end_axis, false);
+    return softmax_layer_generic<T>(src, beta, axis, is_log);
 }
 
 template < typename T, typename std::enable_if < std::is_same<T, uint8_t>::value || std::is_same<T, int8_t>::value, int >::type >
-SimpleTensor<T> softmax_layer(const SimpleTensor<T> &src, float beta, int32_t reduce_end_axis)
+SimpleTensor<T> softmax_layer(const SimpleTensor<T> &src, float beta, int32_t axis, bool is_log)
 {
-    const QuantizationInfo output_quantization_info = arm_compute::get_softmax_output_quantization_info(src.data_type(), false);
+    const QuantizationInfo output_quantization_info = arm_compute::get_softmax_output_quantization_info(src.data_type(), is_log);
 
     SimpleTensor<float> src_tmp = convert_from_asymmetric(src);
-    SimpleTensor<float> dst_tmp = softmax_layer<float>(src_tmp, beta, reduce_end_axis);
+    SimpleTensor<float> dst_tmp = softmax_layer<float>(src_tmp, beta, axis, is_log);
     SimpleTensor<T>     dst     = convert_to_asymmetric<T>(dst_tmp, output_quantization_info);
     return dst;
 }
 
-template SimpleTensor<float> softmax_layer(const SimpleTensor<float> &src, float beta, int32_t reduce_end_axis);
-template SimpleTensor<half> softmax_layer(const SimpleTensor<half> &src, float beta, int32_t reduce_end_axis);
-template SimpleTensor<uint8_t> softmax_layer(const SimpleTensor<uint8_t> &src, float beta, int32_t reduce_end_axis);
-template SimpleTensor<int8_t> softmax_layer(const SimpleTensor<int8_t> &src, float beta, int32_t reduce_end_axis);
+template SimpleTensor<float> softmax_layer(const SimpleTensor<float> &src, float beta, int32_t axis, bool is_log);
+template SimpleTensor<half> softmax_layer(const SimpleTensor<half> &src, float beta, int32_t axis, bool is_log);
+template SimpleTensor<uint8_t> softmax_layer(const SimpleTensor<uint8_t> &src, float beta, int32_t axis, bool is_log);
+template SimpleTensor<int8_t> softmax_layer(const SimpleTensor<int8_t> &src, float beta, int32_t axis, bool is_log);
+
 } // namespace reference
 } // namespace validation
 } // namespace test