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, 3 insertions, 3 deletions

diff --git a/src/runtime/GLES_COMPUTE/functions/GCSoftmaxLayer.cpp b/src/runtime/GLES_COMPUTE/functions/GCSoftmaxLayer.cpp
index 48d8cb576b..fdb9a42f13 100644
--- a/src/runtime/GLES_COMPUTE/functions/GCSoftmaxLayer.cpp
+++ b/src/runtime/GLES_COMPUTE/functions/GCSoftmaxLayer.cpp
@@ -34,13 +34,13 @@ GCSoftmaxLayer::GCSoftmaxLayer(std::shared_ptr<IMemoryManager> memory_manager)
 {
 }
 
-void GCSoftmaxLayer::configure(const IGCTensor *input, IGCTensor *output, float beta, size_t reduce_end_axis)
+void GCSoftmaxLayer::configure(const IGCTensor *input, IGCTensor *output, float beta, int32_t axis)
 {
-    ARM_COMPUTE_UNUSED(beta, reduce_end_axis);
+    ARM_COMPUTE_UNUSED(beta, axis);
 
     ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(input, 1, DataType::F16, DataType::F32);
     ARM_COMPUTE_ERROR_ON(beta != 1.0f);
-    ARM_COMPUTE_ERROR_ON_MSG(reduce_end_axis != 0, "Reduce_end_axis must be 0 for GLES");
+    ARM_COMPUTE_ERROR_ON_MSG(axis != 0, "axis must be 0 for GLES");
 
     // Create intermediate tensors shapes
     _tmp.allocator()->init(TensorInfo(input->info()->tensor_shape(), input->info()->num_channels(), input->info()->data_type()));