implement balanced tree reduce for xilinxhls backend

Author: francescobrivio

conifer/backends/xilinxhls/firmware/BDT_unrolled.h

@@ -5,8 +5,37 @@
 
 namespace BDT{
 
+/* ---
+* Balanced tree reduce implementation.
+* Reduces an array of inputs to a single value using the template binary operator 'Op',
+* for example summing all elements with OpAdd, or finding the maximum with OpMax
+* Use only when the input array is fully unrolled. Or, slice out a fully unrolled section
+* before applying and accumulate the result over the rolled dimension.
+* Required for emulation to guarantee equality of ordering.
+* --- */
+constexpr int floorlog2(int x) { return (x < 2) ? 0 : 1 + floorlog2(x / 2); }
+
+constexpr int pow2(int x) { return x == 0 ? 1 : 2 * pow2(x - 1); }
+
+template <class T, int N, class Op> T reduce(const T *x, Op op) {
+  static constexpr int leftN = pow2(floorlog2(N - 1)) > 0 ? pow2(floorlog2(N - 1)) : 0;
+  static constexpr int rightN = N - leftN > 0 ? N - leftN : 0;
+  if (N == 1) {
+    return x[0];
+  }
+  if (N == 2) {
+    return op(x[0], x[1]);
+  }
+  return op(reduce<T, leftN, Op>(x, op), reduce<T, rightN, Op>(x + leftN, op));
+}
+
+template <class T> class OpAdd {
+  public:
+    T operator()(T a, T b) { return a + b; }
+};
+
+// Number of trees given number of classes
 constexpr int fn_classes(int n_classes){
-  // Number of trees given number of classes
   return n_classes == 2 ? 1 : n_classes;
 }
 
@@ -99,23 +128,24 @@ struct BDT{
 public:
   score_t normalisation;
   score_t init_predict[fn_classes(n_classes)];
+  OpAdd<score_t> op_add;
 
-  void tree_scores(input_t x, score_t scores[n_trees][fn_classes(n_classes)]) const;
+  void tree_scores(input_t x, score_t scores[fn_classes(n_classes)][n_trees]) const;
 
   void decision_function(input_t x, score_t score[fn_classes(n_classes)]) const{
-    score_t scores[n_trees][fn_classes(n_classes)];
+    score_t scores[fn_classes(n_classes)][n_trees];
     #pragma HLS ARRAY_PARTITION variable=scores dim=0
+    // Get predictions scores
+    tree_scores(x, scores);
+    // Reduce
+    Reduce:
     for(int j = 0; j < fn_classes(n_classes); j++){
+      // Init predictions
       score[j] = init_predict[j];
+      // Sum predictions from trees via "reduce" method
+      score[j] += reduce<score_t, n_trees, OpAdd<score_t>>(scores[j], op_add);
     }
-    tree_scores(x, scores);
-    Trees:
-    for(int i = 0; i < n_trees; i++){
-      Classes:
-      for(int j = 0; j < fn_classes(n_classes); j++){
-        score[j] += scores[i][j];
-      }
-    }
+    // Normalize predictions
     for(int j = 0; j < fn_classes(n_classes); j++){
       score[j] *= normalisation;
     }

conifer/backends/xilinxhls/hls-template/firmware/BDT_unrolled.cpp

@@ -2,7 +2,7 @@
 #include "parameters.h"
 
 template<>
-void BDT::BDT<n_trees, n_classes, input_arr_t, score_t, threshold_t>::tree_scores(input_arr_t x, score_t scores[n_trees][fn_classes(n_classes)]) const {
+void BDT::BDT<n_trees, n_classes, input_arr_t, score_t, threshold_t>::tree_scores(input_arr_t x, score_t scores[fn_classes(n_classes)][n_trees]) const {
   // conifer insert tree_scores
 }
 

conifer/backends/xilinxhls/writer.py

@@ -139,7 +139,7 @@ def write_bdt_h(self):
                     newline = ''
                     for it, trees in enumerate(self.trees):
                         for ic, tree in enumerate(trees):
-                            newline += f'  scores[{it}][{ic}] = tree_{it}_{ic}.decision_function(x);\n'
+                            newline += f'  scores[{ic}][{it}] = tree_{ic}_{it}.decision_function(x);\n'
                 else:
                     newline = line
                 fout.write(newline)
@@ -227,7 +227,7 @@ def _write_parameters_h_unrolled(self, fout):
             for iclass, tree in enumerate(trees):
                 fout.write(f'static const BDT::Tree<{itree*nc+iclass}, {tree.n_nodes()}, {tree.n_leaves()}')
                 fout.write(f', input_arr_t, score_t, threshold_t>')
-                fout.write(f' tree_{itree}_{iclass} = {{\n')
+                fout.write(f' tree_{iclass}_{itree} = {{\n')
                 # loop over fields
                 for ifield, field in enumerate(tree_fields):
                     newline = '    {'