added feature infos to JSON dump (#2660)

* added feature infos to JSON dump

* slight json schema refactor

* simpified code

* refactor feature_infos

* refactoring

* Update src/boosting/gbdt.cpp

* Update dataset.h

* Update include/LightGBM/dataset.h

* simplify

* Apply suggestions from code review

* parse string and construct JSON objs
Co-authored-by: Guolin Ke <guolin.ke@outlook.com>

include/LightGBM/bin.h

@@ -178,7 +178,7 @@ class BinMapper {
   /*!
   * \brief Get bin info
   */
-  inline std::string bin_info() const {
+  inline std::string bin_info_string() const {
     if (bin_type_ == BinType::CategoricalBin) {
       return Common::Join(bin_2_categorical_, ":");
     } else {

include/LightGBM/dataset.h

@@ -616,13 +616,13 @@ class Dataset {
 
   inline std::vector<std::string> feature_infos() const {
     std::vector<std::string> bufs;
-    for (int i = 0; i < num_total_features_; i++) {
+    for (int i = 0; i < num_total_features_; ++i) {
       int fidx = used_feature_map_[i];
-      if (fidx == -1) {
+      if (fidx < 0) {
         bufs.push_back("none");
       } else {
         const auto bin_mapper = FeatureBinMapper(fidx);
-        bufs.push_back(bin_mapper->bin_info());
+        bufs.push_back(bin_mapper->bin_info_string());
       }
     }
     return bufs;

include/LightGBM/utils/array_args.h

@@ -195,4 +195,3 @@ class ArrayArgs {
 }  // namespace LightGBM
 
 #endif   // LightGBM_UTILS_ARRAY_AGRS_H_
-

src/boosting/gbdt_model_text.cpp

@@ -5,6 +5,7 @@
 #include <LightGBM/config.h>
 #include <LightGBM/metric.h>
 #include <LightGBM/objective_function.h>
+#include <LightGBM/utils/array_args.h>
 #include <LightGBM/utils/common.h>
 
 #include <string>
@@ -39,6 +40,36 @@ std::string GBDT::DumpModel(int start_iteration, int num_iteration) const {
   str_buf << "\"monotone_constraints\":["
           << Common::Join(monotone_constraints_, ",") << "]," << '\n';
 
+  str_buf << "\"feature_infos\":" << "{";
+  bool first_obj = true;
+  for (size_t i = 0; i < feature_infos_.size(); ++i) {
+    std::stringstream json_str_buf;
+    auto strs = Common::Split(feature_infos_[i].c_str(), ":");
+    if (strs[0][0] == '[') {
+      strs[0].erase(0, 1);  // remove '['
+      strs[1].erase(strs[1].size() - 1);  // remove ']'
+      json_str_buf << "{\"min_value\":" << strs[0] << ",";
+      json_str_buf << "\"max_value\":" << strs[1] << ",";
+      json_str_buf << "\"values\":[]}";
+    } else if (strs[0] != "none") {  // categorical feature
+      auto vals = Common::StringToArray<int>(feature_infos_[i], ':');
+      auto max_idx = ArrayArgs<int>::ArgMax(vals);
+      auto min_idx = ArrayArgs<int>::ArgMin(vals);
+      json_str_buf << "{\"min_value\":" << vals[min_idx] << ",";
+      json_str_buf << "\"max_value\":" << vals[max_idx] << ",";
+      json_str_buf << "\"values\":[" << Common::Join(vals, ",") << "]}";
+    } else {  // unused feature
+      continue;
+    }
+    if (!first_obj) {
+      str_buf << ",";
+    }
+    str_buf << "\"" << feature_names_[i] << "\":";
+    str_buf << json_str_buf.str();
+    first_obj = false;
+  }
+  str_buf << "}," << '\n';
+
   str_buf << "\"tree_info\":[";
   int num_used_model = static_cast<int>(models_.size());
   int total_iteration = num_used_model / num_tree_per_iteration_;
@@ -70,12 +101,11 @@ std::string GBDT::DumpModel(int start_iteration, int num_iteration) const {
     }
   }
   str_buf << '\n' << "\"feature_importances\":" << "{";
-  if (!pairs.empty()) {
-    str_buf << "\"" << pairs[0].second << "\":" << std::to_string(pairs[0].first);
-    for (size_t i = 1; i < pairs.size(); ++i) {
+  for (size_t i = 0; i < pairs.size(); ++i) {
+    if (i > 0) {
       str_buf << ",";
-      str_buf << "\"" << pairs[i].second << "\":" << std::to_string(pairs[i].first);
     }
+    str_buf << "\"" << pairs[i].second << "\":" << std::to_string(pairs[i].first);
   }
   str_buf << "}" << '\n';