examples: add a new accumulator object for statistics computations

Author: jcelerier

examples/Advanced/Utilities/Accumulator.hpp

@@ -0,0 +1,169 @@
+#pragma once
+
+#include <boost/accumulators/accumulators.hpp>
+#include <boost/accumulators/statistics.hpp>
+#include <boost/accumulators/statistics/count.hpp>
+#include <boost/accumulators/statistics/kurtosis.hpp>
+#include <boost/accumulators/statistics/mean.hpp>
+#include <boost/accumulators/statistics/median.hpp>
+#include <boost/accumulators/statistics/sum.hpp>
+#include <boost/accumulators/statistics/variance.hpp>
+#include <halp/controls.hpp>
+#include <halp/mappers.hpp>
+#include <halp/meta.hpp>
+#include <ossia/detail/math.hpp>
+
+namespace ao
+{
+namespace ba = boost::accumulators;
+namespace bt = ba::tag;
+/**
+ * @brief Calibrate a value and output it between 0-1 according to
+ * the range of inputs
+ */
+struct Accumulator
+{
+public:
+  halp_meta(name, "Accumulator")
+  halp_meta(c_name, "accumulator")
+  halp_meta(category, "Control/Mappings")
+  halp_meta(author, "Jean-Michaël Celerier")
+  halp_meta(description, "Accumulate statistics about incoming values")
+  halp_meta(manual_url, "https://ossia.io/score-docs/processes/accumulator.html")
+  halp_meta(uuid, "5c5b37b5-da06-432a-bc51-81657b6d59e1")
+
+  struct inputs_t
+  {
+    halp::val_port<"In", std::optional<float>> in;
+    struct : halp::impulse_button<"Reset">
+    {
+      void update(Accumulator& self)
+      {
+        std::destroy_at(&self.minmax);
+        std::construct_at(&self.minmax);
+      }
+    } reset;
+  } inputs;
+
+  struct
+  {
+    struct : halp::val_port<"Count", float>
+    {
+      struct range
+      {
+        const float min = 0.f;
+        const float max = 1.f;
+        const float init = 0.f;
+      };
+    } count;
+    struct : halp::val_port<"Sum", float>
+    {
+      struct range
+      {
+        const float min = 0.f;
+        const float max = 1.f;
+        const float init = 0.f;
+      };
+    } sum;
+    struct : halp::val_port<"Consecutive difference", float>
+    {
+      struct range
+      {
+        const float min = 0.f;
+        const float max = 1.f;
+        const float init = 0.f;
+      };
+    } diff;
+    struct : halp::val_port<"Mean", float>
+    {
+      struct range
+      {
+        const float min = 1.f;
+        const float max = 0.f;
+        const float init = 0.f;
+      };
+    } mean;
+    struct : halp::val_port<"Variance", float>
+    {
+      struct range
+      {
+        const float min = 0.f;
+        const float max = 1.f;
+        const float init = 0.f;
+      };
+    } variance;
+    struct : halp::val_port<"Median", float>
+    {
+      struct range
+      {
+        const float min = 1.f;
+        const float max = 0.f;
+        const float init = 0.f;
+      };
+    } median;
+    struct : halp::val_port<"Kurtosis", float>
+    {
+      struct range
+      {
+        const float min = 1.f;
+        const float max = 0.f;
+        const float init = 0.f;
+      };
+    } kurtosis;
+
+    struct : halp::val_port<"Min", float>
+    {
+      struct range
+      {
+        const float min = 0.f;
+        const float max = 1.f;
+        const float init = 0.f;
+      };
+    } min;
+    struct : halp::val_port<"Max", float>
+    {
+      struct range
+      {
+        const float min = 0.f;
+        const float max = 1.f;
+        const float init = 0.f;
+      };
+    } max;
+
+  } outputs;
+
+  using accum = ba::accumulator_set<
+      float, ba::stats<
+                 ba::tag::count, ba::tag::sum, ba::tag::min, ba::tag::max, ba::tag::mean,
+                 ba::tag::variance, ba::tag::median, ba::tag::kurtosis>>;
+
+  accum minmax{};
+  float consecutive_difference{};
+  bool consecutive_difference_sign{};
+
+  void operator()() noexcept
+  {
+    using namespace ba;
+    if(inputs.in.value)
+    {
+      float v = *inputs.in.value;
+      this->minmax(v);
+      if(consecutive_difference_sign ^= true)
+        consecutive_difference += v;
+      else
+        consecutive_difference -= v;
+
+      outputs.count.value = ba::extract::count(minmax);
+      outputs.sum.value = ba::extract::sum(minmax);
+      outputs.diff.value = consecutive_difference;
+      outputs.min.value = ba::extract::min(minmax);
+      outputs.max.value = ba::extract::max(minmax);
+      outputs.mean.value = ba::extract::mean(minmax);
+      outputs.variance.value = ba::extract::variance(minmax);
+      outputs.median.value = ba::extract::median(minmax);
+      outputs.kurtosis.value = ba::extract::kurtosis(minmax);
+    }
+  }
+};
+
+}