ELU layer with basic tests

1 files changed, 86 insertions, 0 deletions

diff --git a/include/caffe/layers/elu_layer.hpp b/include/caffe/layers/elu_layer.hpp
new file mode 100644
index 00000000..0796e898
--- /dev/null
+++ b/include/caffe/layers/elu_layer.hpp
@@ -0,0 +1,86 @@
+#ifndef CAFFE_ELU_LAYER_HPP_
+#define CAFFE_ELU_LAYER_HPP_
+
+#include <vector>
+
+#include "caffe/blob.hpp"
+#include "caffe/layer.hpp"
+#include "caffe/proto/caffe.pb.h"
+
+#include "caffe/layers/neuron_layer.hpp"
+
+namespace caffe {
+
+/**
+ * @brief Exponential Linear Unit non-linearity @f$
+ *        y = \left\{
+ *        \begin{array}{lr}
+ *            x                  & \mathrm{if} \; x > 0 \\
+ *            \alpha (\exp(x)-1) & \mathrm{if} \; x \le 0
+ *        \end{array} \right.
+ *      @f$.  
+ */
+template <typename Dtype>
+class ELULayer : public NeuronLayer<Dtype> {
+ public:
+  /**
+   * @param param provides ELUParameter elu_param,
+   *     with ELULayer options:
+   *   - alpha (\b optional, default 1).
+   *     the value @f$ \alpha @f$ by which controls saturation for negative inputs.
+   */
+  explicit ELULayer(const LayerParameter& param)
+      : NeuronLayer<Dtype>(param) {}
+
+  virtual inline const char* type() const { return "ELU"; }
+
+ protected:
+  /**
+   * @param bottom input Blob vector (length 1)
+   *   -# @f$ (N \times C \times H \times W) @f$
+   *      the inputs @f$ x @f$
+   * @param top output Blob vector (length 1)
+   *   -# @f$ (N \times C \times H \times W) @f$
+   *      the computed outputs @f$
+   *        y = \left\{
+   *        \begin{array}{lr}
+   *            x                  & \mathrm{if} \; x > 0 \\
+   *            \alpha (\exp(x)-1) & \mathrm{if} \; x \le 0
+   *        \end{array} \right.
+   *      @f$.  
+   */
+  virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+      const vector<Blob<Dtype>*>& top);
+  virtual void Forward_gpu(const vector<Blob<Dtype>*>& bottom,
+      const vector<Blob<Dtype>*>& top);
+
+  /**
+   * @brief Computes the error gradient w.r.t. the ELU inputs.
+   *
+   * @param top output Blob vector (length 1), providing the error gradient with
+   *      respect to the outputs
+   *   -# @f$ (N \times C \times H \times W) @f$
+   *      containing error gradients @f$ \frac{\partial E}{\partial y} @f$
+   *      with respect to computed outputs @f$ y @f$
+   * @param propagate_down see Layer::Backward.
+   * @param bottom input Blob vector (length 1)
+   *   -# @f$ (N \times C \times H \times W) @f$
+   *      the inputs @f$ x @f$; Backward fills their diff with
+   *      gradients @f$
+   *        \frac{\partial E}{\partial x} = \left\{
+   *        \begin{array}{lr}
+   *            1           & \mathrm{if} \; x > 0 \\
+   *            y + \alpha  & \mathrm{if} \; x \le 0
+   *        \end{array} \right.
+   *      @f$ if propagate_down[0].
+   */
+  virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
+      const vector<bool>& propagate_down, const vector<Blob<Dtype>*>& bottom);
+  virtual void Backward_gpu(const vector<Blob<Dtype>*>& top,
+      const vector<bool>& propagate_down, const vector<Blob<Dtype>*>& bottom);
+};
+
+
+}  // namespace caffe
+
+#endif  // CAFFE_ELU_LAYER_HPP_