Add support for importing LightGBM models

11 files changed, 994 insertions, 6 deletions

diff --git a/config-model/src/main/java/com/yahoo/searchdefinition/expressiontransforms/ExpressionTransforms.java b/config-model/src/main/java/com/yahoo/searchdefinition/expressiontransforms/ExpressionTransforms.java
index 6fdf448a39b..a6707ec7ac0 100644
--- a/config-model/src/main/java/com/yahoo/searchdefinition/expressiontransforms/ExpressionTransforms.java
+++ b/config-model/src/main/java/com/yahoo/searchdefinition/expressiontransforms/ExpressionTransforms.java
@@ -27,6 +27,7 @@ public class ExpressionTransforms {
                 ImmutableList.of(new TensorFlowFeatureConverter(),
                                  new OnnxFeatureConverter(),
                                  new XgboostFeatureConverter(),
+                                 new LightGBMFeatureConverter(),
                                  new ConstantDereferencer(),
                                  new ConstantTensorTransformer(),
                                  new FunctionInliner(),
diff --git a/config-model/src/main/java/com/yahoo/searchdefinition/expressiontransforms/LightGBMFeatureConverter.java b/config-model/src/main/java/com/yahoo/searchdefinition/expressiontransforms/LightGBMFeatureConverter.java
new file mode 100644
index 00000000000..5bde627dc0a
--- /dev/null
+++ b/config-model/src/main/java/com/yahoo/searchdefinition/expressiontransforms/LightGBMFeatureConverter.java
@@ -0,0 +1,59 @@
+// Copyright 2020 Oath Inc. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root.
+package com.yahoo.searchdefinition.expressiontransforms;
+
+import com.yahoo.path.Path;
+import com.yahoo.searchlib.rankingexpression.rule.Arguments;
+import com.yahoo.searchlib.rankingexpression.rule.CompositeNode;
+import com.yahoo.searchlib.rankingexpression.rule.ExpressionNode;
+import com.yahoo.searchlib.rankingexpression.rule.ReferenceNode;
+import com.yahoo.searchlib.rankingexpression.transform.ExpressionTransformer;
+import com.yahoo.vespa.model.ml.ConvertedModel;
+import com.yahoo.vespa.model.ml.FeatureArguments;
+
+import java.io.UncheckedIOException;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * Replaces instances of the lightgbm(model-path) pseudofeature with the
+ * native Vespa ranking expression implementing the same computation.
+ *
+ * @author lesters
+ */
+public class LightGBMFeatureConverter extends ExpressionTransformer<RankProfileTransformContext> {
+
+    /** A cache of imported models indexed by model path. This avoids importing the same model multiple times. */
+    private final Map<Path, ConvertedModel> convertedLightGBMModels = new HashMap<>();
+
+    @Override
+    public ExpressionNode transform(ExpressionNode node, RankProfileTransformContext context) {
+        if (node instanceof ReferenceNode)
+            return transformFeature((ReferenceNode) node, context);
+        else if (node instanceof CompositeNode)
+            return super.transformChildren((CompositeNode) node, context);
+        else
+            return node;
+    }
+
+    private ExpressionNode transformFeature(ReferenceNode feature, RankProfileTransformContext context) {
+        if ( ! feature.getName().equals("lightgbm")) return feature;
+
+        try {
+            FeatureArguments arguments = asFeatureArguments(feature.getArguments());
+            ConvertedModel convertedModel =
+                    convertedLightGBMModels.computeIfAbsent(arguments.path(),
+                                                           path -> ConvertedModel.fromSourceOrStore(path, true, context));
+            return convertedModel.expression(arguments, context);
+        } catch (IllegalArgumentException | UncheckedIOException e) {
+            throw new IllegalArgumentException("Could not use LightGBM model from " + feature, e);
+        }
+    }
+
+    private FeatureArguments asFeatureArguments(Arguments arguments) {
+        if (arguments.size() != 1)
+            throw new IllegalArgumentException("A lightgbm node must take a single argument pointing to " +
+                                               "the LightGBM model file under [application]/models");
+        return new FeatureArguments(arguments);
+    }
+
+}
diff --git a/config-model/src/test/cfg/application/ml_models/models/lightgbm_regression.json b/config-model/src/test/cfg/application/ml_models/models/lightgbm_regression.json
new file mode 100644
index 00000000000..cf0488ecd8b
--- /dev/null
+++ b/config-model/src/test/cfg/application/ml_models/models/lightgbm_regression.json
@@ -0,0 +1,275 @@
+{
+  "name": "tree",
+  "version": "v3",
+  "num_class": 1,
+  "num_tree_per_iteration": 1,
+  "label_index": 0,
+  "max_feature_idx": 3,
+  "average_output": false,
+  "objective": "regression",
+  "feature_names": [
+    "numerical_1",
+    "numerical_2",
+    "categorical_1",
+    "categorical_2"
+  ],
+  "monotone_constraints": [],
+  "tree_info": [
+    {
+      "tree_index": 0,
+      "num_leaves": 3,
+      "num_cat": 1,
+      "shrinkage": 1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 1,
+        "split_gain": 68.5353012084961,
+        "threshold": 0.46643291586559305,
+        "decision_type": "<=",
+        "default_left": true,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "leaf_index": 0,
+          "leaf_value": 2.1594397038037663,
+          "leaf_weight": 469,
+          "leaf_count": 469
+        },
+        "right_child": {
+          "split_index": 1,
+          "split_feature": 3,
+          "split_gain": 41.27640151977539,
+          "threshold": "2||3||4",
+          "decision_type": "==",
+          "default_left": false,
+          "missing_type": "NaN",
+          "internal_value": 0.246035,
+          "internal_weight": 531,
+          "internal_count": 531,
+          "left_child": {
+            "leaf_index": 1,
+            "leaf_value": 2.235297305276056,
+            "leaf_weight": 302,
+            "leaf_count": 302
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": 2.1792953471546546,
+            "leaf_weight": 229,
+            "leaf_count": 229
+          }
+        }
+      }
+    },
+    {
+      "tree_index": 1,
+      "num_leaves": 3,
+      "num_cat": 1,
+      "shrinkage": 0.1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 2,
+        "split_gain": 64.22250366210938,
+        "threshold": "3||4",
+        "decision_type": "==",
+        "default_left": false,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "leaf_index": 0,
+          "leaf_value": 0.03070842919354316,
+          "leaf_weight": 399,
+          "leaf_count": 399
+        },
+        "right_child": {
+          "split_index": 1,
+          "split_feature": 0,
+          "split_gain": 36.74250030517578,
+          "threshold": 0.5102250691730842,
+          "decision_type": "<=",
+          "default_left": true,
+          "missing_type": "NaN",
+          "internal_value": -0.204906,
+          "internal_weight": 601,
+          "internal_count": 601,
+          "left_child": {
+            "leaf_index": 1,
+            "leaf_value": -0.04439151147520909,
+            "leaf_weight": 315,
+            "leaf_count": 315
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": 0.005117411709368601,
+            "leaf_weight": 286,
+            "leaf_count": 286
+          }
+        }
+      }
+    },
+    {
+      "tree_index": 2,
+      "num_leaves": 3,
+      "num_cat": 0,
+      "shrinkage": 0.1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 1,
+        "split_gain": 57.1327018737793,
+        "threshold": 0.668665477622446,
+        "decision_type": "<=",
+        "default_left": true,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "split_index": 1,
+          "split_feature": 1,
+          "split_gain": 40.859100341796875,
+          "threshold": 0.008118820676863816,
+          "decision_type": "<=",
+          "default_left": true,
+          "missing_type": "NaN",
+          "internal_value": -0.162926,
+          "internal_weight": 681,
+          "internal_count": 681,
+          "left_child": {
+            "leaf_index": 0,
+            "leaf_value": -0.15361238490967524,
+            "leaf_weight": 21,
+            "leaf_count": 21
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": -0.01192330846157292,
+            "leaf_weight": 660,
+            "leaf_count": 660
+          }
+        },
+        "right_child": {
+          "leaf_index": 1,
+          "leaf_value": 0.03499044894987518,
+          "leaf_weight": 319,
+          "leaf_count": 319
+        }
+      }
+    },
+    {
+      "tree_index": 3,
+      "num_leaves": 3,
+      "num_cat": 1,
+      "shrinkage": 0.1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 0,
+        "split_gain": 54.77090072631836,
+        "threshold": 0.5201391072644542,
+        "decision_type": "<=",
+        "default_left": true,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "leaf_index": 0,
+          "leaf_value": -0.02141000620783247,
+          "leaf_weight": 543,
+          "leaf_count": 543
+        },
+        "right_child": {
+          "split_index": 1,
+          "split_feature": 2,
+          "split_gain": 27.200700759887695,
+          "threshold": "0||1",
+          "decision_type": "==",
+          "default_left": false,
+          "missing_type": "NaN",
+          "internal_value": 0.255704,
+          "internal_weight": 457,
+          "internal_count": 457,
+          "left_child": {
+            "leaf_index": 1,
+            "leaf_value": -0.004121485787596721,
+            "leaf_weight": 191,
+            "leaf_count": 191
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": 0.04534090904886873,
+            "leaf_weight": 266,
+            "leaf_count": 266
+          }
+        }
+      }
+    },
+    {
+      "tree_index": 4,
+      "num_leaves": 3,
+      "num_cat": 1,
+      "shrinkage": 0.1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 3,
+        "split_gain": 51.84349822998047,
+        "threshold": "2||3||4",
+        "decision_type": "==",
+        "default_left": false,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "split_index": 1,
+          "split_feature": 1,
+          "split_gain": 39.352699279785156,
+          "threshold": 0.27283279016959255,
+          "decision_type": "<=",
+          "default_left": true,
+          "missing_type": "NaN",
+          "internal_value": 0.188414,
+          "internal_weight": 593,
+          "internal_count": 593,
+          "left_child": {
+            "leaf_index": 0,
+            "leaf_value": -0.01924803254356527,
+            "leaf_weight": 184,
+            "leaf_count": 184
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": 0.03643772842347651,
+            "leaf_weight": 409,
+            "leaf_count": 409
+          }
+        },
+        "right_child": {
+          "leaf_index": 1,
+          "leaf_value": -0.02701711918923075,
+          "leaf_weight": 407,
+          "leaf_count": 407
+        }
+      }
+    }
+  ],
+  "pandas_categorical": [
+    [
+      "a",
+      "b",
+      "c",
+      "d",
+      "e"
+    ],
+    [
+      "i",
+      "j",
+      "k",
+      "l",
+      "m"
+    ]
+  ]
+}
\ No newline at end of file
diff --git a/config-model/src/test/cfg/application/ml_models/searchdefinitions/test.sd b/config-model/src/test/cfg/application/ml_models/searchdefinitions/test.sd
index ab5e42f983d..247df8a0241 100644
--- a/config-model/src/test/cfg/application/ml_models/searchdefinitions/test.sd
+++ b/config-model/src/test/cfg/application/ml_models/searchdefinitions/test.sd
@@ -33,8 +33,12 @@ search test {
             expression: xgboost("xgboost_2_2")
         }
 
+        function my_lightgbm() {
+            expression: lightgbm("lightgbm_regression")
+        }
+
         first-phase {
-            expression: mnist_tensorflow + mnist_softmax_tensorflow + mnist_softmax_onnx + my_xgboost
+            expression: mnist_tensorflow + mnist_softmax_tensorflow + mnist_softmax_onnx + my_xgboost + my_lightgbm
         }
 
     }
diff --git a/config-model/src/test/cfg/application/ml_serving/models/lightgbm_regression.json b/config-model/src/test/cfg/application/ml_serving/models/lightgbm_regression.json
new file mode 100644
index 00000000000..cf0488ecd8b
--- /dev/null
+++ b/config-model/src/test/cfg/application/ml_serving/models/lightgbm_regression.json
@@ -0,0 +1,275 @@
+{
+  "name": "tree",
+  "version": "v3",
+  "num_class": 1,
+  "num_tree_per_iteration": 1,
+  "label_index": 0,
+  "max_feature_idx": 3,
+  "average_output": false,
+  "objective": "regression",
+  "feature_names": [
+    "numerical_1",
+    "numerical_2",
+    "categorical_1",
+    "categorical_2"
+  ],
+  "monotone_constraints": [],
+  "tree_info": [
+    {
+      "tree_index": 0,
+      "num_leaves": 3,
+      "num_cat": 1,
+      "shrinkage": 1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 1,
+        "split_gain": 68.5353012084961,
+        "threshold": 0.46643291586559305,
+        "decision_type": "<=",
+        "default_left": true,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "leaf_index": 0,
+          "leaf_value": 2.1594397038037663,
+          "leaf_weight": 469,
+          "leaf_count": 469
+        },
+        "right_child": {
+          "split_index": 1,
+          "split_feature": 3,
+          "split_gain": 41.27640151977539,
+          "threshold": "2||3||4",
+          "decision_type": "==",
+          "default_left": false,
+          "missing_type": "NaN",
+          "internal_value": 0.246035,
+          "internal_weight": 531,
+          "internal_count": 531,
+          "left_child": {
+            "leaf_index": 1,
+            "leaf_value": 2.235297305276056,
+            "leaf_weight": 302,
+            "leaf_count": 302
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": 2.1792953471546546,
+            "leaf_weight": 229,
+            "leaf_count": 229
+          }
+        }
+      }
+    },
+    {
+      "tree_index": 1,
+      "num_leaves": 3,
+      "num_cat": 1,
+      "shrinkage": 0.1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 2,
+        "split_gain": 64.22250366210938,
+        "threshold": "3||4",
+        "decision_type": "==",
+        "default_left": false,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "leaf_index": 0,
+          "leaf_value": 0.03070842919354316,
+          "leaf_weight": 399,
+          "leaf_count": 399
+        },
+        "right_child": {
+          "split_index": 1,
+          "split_feature": 0,
+          "split_gain": 36.74250030517578,
+          "threshold": 0.5102250691730842,
+          "decision_type": "<=",
+          "default_left": true,
+          "missing_type": "NaN",
+          "internal_value": -0.204906,
+          "internal_weight": 601,
+          "internal_count": 601,
+          "left_child": {
+            "leaf_index": 1,
+            "leaf_value": -0.04439151147520909,
+            "leaf_weight": 315,
+            "leaf_count": 315
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": 0.005117411709368601,
+            "leaf_weight": 286,
+            "leaf_count": 286
+          }
+        }
+      }
+    },
+    {
+      "tree_index": 2,
+      "num_leaves": 3,
+      "num_cat": 0,
+      "shrinkage": 0.1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 1,
+        "split_gain": 57.1327018737793,
+        "threshold": 0.668665477622446,
+        "decision_type": "<=",
+        "default_left": true,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "split_index": 1,
+          "split_feature": 1,
+          "split_gain": 40.859100341796875,
+          "threshold": 0.008118820676863816,
+          "decision_type": "<=",
+          "default_left": true,
+          "missing_type": "NaN",
+          "internal_value": -0.162926,
+          "internal_weight": 681,
+          "internal_count": 681,
+          "left_child": {
+            "leaf_index": 0,
+            "leaf_value": -0.15361238490967524,
+            "leaf_weight": 21,
+            "leaf_count": 21
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": -0.01192330846157292,
+            "leaf_weight": 660,
+            "leaf_count": 660
+          }
+        },
+        "right_child": {
+          "leaf_index": 1,
+          "leaf_value": 0.03499044894987518,
+          "leaf_weight": 319,
+          "leaf_count": 319
+        }
+      }
+    },
+    {
+      "tree_index": 3,
+      "num_leaves": 3,
+      "num_cat": 1,
+      "shrinkage": 0.1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 0,
+        "split_gain": 54.77090072631836,
+        "threshold": 0.5201391072644542,
+        "decision_type": "<=",
+        "default_left": true,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "leaf_index": 0,
+          "leaf_value": -0.02141000620783247,
+          "leaf_weight": 543,
+          "leaf_count": 543
+        },
+        "right_child": {
+          "split_index": 1,
+          "split_feature": 2,
+          "split_gain": 27.200700759887695,
+          "threshold": "0||1",
+          "decision_type": "==",
+          "default_left": false,
+          "missing_type": "NaN",
+          "internal_value": 0.255704,
+          "internal_weight": 457,
+          "internal_count": 457,
+          "left_child": {
+            "leaf_index": 1,
+            "leaf_value": -0.004121485787596721,
+            "leaf_weight": 191,
+            "leaf_count": 191
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": 0.04534090904886873,
+            "leaf_weight": 266,
+            "leaf_count": 266
+          }
+        }
+      }
+    },
+    {
+      "tree_index": 4,
+      "num_leaves": 3,
+      "num_cat": 1,
+      "shrinkage": 0.1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 3,
+        "split_gain": 51.84349822998047,
+        "threshold": "2||3||4",
+        "decision_type": "==",
+        "default_left": false,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "split_index": 1,
+          "split_feature": 1,
+          "split_gain": 39.352699279785156,
+          "threshold": 0.27283279016959255,
+          "decision_type": "<=",
+          "default_left": true,
+          "missing_type": "NaN",
+          "internal_value": 0.188414,
+          "internal_weight": 593,
+          "internal_count": 593,
+          "left_child": {
+            "leaf_index": 0,
+            "leaf_value": -0.01924803254356527,
+            "leaf_weight": 184,
+            "leaf_count": 184
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": 0.03643772842347651,
+            "leaf_weight": 409,
+            "leaf_count": 409
+          }
+        },
+        "right_child": {
+          "leaf_index": 1,
+          "leaf_value": -0.02701711918923075,
+          "leaf_weight": 407,
+          "leaf_count": 407
+        }
+      }
+    }
+  ],
+  "pandas_categorical": [
+    [
+      "a",
+      "b",
+      "c",
+      "d",
+      "e"
+    ],
+    [
+      "i",
+      "j",
+      "k",
+      "l",
+      "m"
+    ]
+  ]
+}
\ No newline at end of file
diff --git a/config-model/src/test/integration/lightgbm/models/regression.json b/config-model/src/test/integration/lightgbm/models/regression.json
new file mode 100644
index 00000000000..cf0488ecd8b
--- /dev/null
+++ b/config-model/src/test/integration/lightgbm/models/regression.json
@@ -0,0 +1,275 @@
+{
+  "name": "tree",
+  "version": "v3",
+  "num_class": 1,
+  "num_tree_per_iteration": 1,
+  "label_index": 0,
+  "max_feature_idx": 3,
+  "average_output": false,
+  "objective": "regression",
+  "feature_names": [
+    "numerical_1",
+    "numerical_2",
+    "categorical_1",
+    "categorical_2"
+  ],
+  "monotone_constraints": [],
+  "tree_info": [
+    {
+      "tree_index": 0,
+      "num_leaves": 3,
+      "num_cat": 1,
+      "shrinkage": 1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 1,
+        "split_gain": 68.5353012084961,
+        "threshold": 0.46643291586559305,
+        "decision_type": "<=",
+        "default_left": true,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "leaf_index": 0,
+          "leaf_value": 2.1594397038037663,
+          "leaf_weight": 469,
+          "leaf_count": 469
+        },
+        "right_child": {
+          "split_index": 1,
+          "split_feature": 3,
+          "split_gain": 41.27640151977539,
+          "threshold": "2||3||4",
+          "decision_type": "==",
+          "default_left": false,
+          "missing_type": "NaN",
+          "internal_value": 0.246035,
+          "internal_weight": 531,
+          "internal_count": 531,
+          "left_child": {
+            "leaf_index": 1,
+            "leaf_value": 2.235297305276056,
+            "leaf_weight": 302,
+            "leaf_count": 302
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": 2.1792953471546546,
+            "leaf_weight": 229,
+            "leaf_count": 229
+          }
+        }
+      }
+    },
+    {
+      "tree_index": 1,
+      "num_leaves": 3,
+      "num_cat": 1,
+      "shrinkage": 0.1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 2,
+        "split_gain": 64.22250366210938,
+        "threshold": "3||4",
+        "decision_type": "==",
+        "default_left": false,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "leaf_index": 0,
+          "leaf_value": 0.03070842919354316,
+          "leaf_weight": 399,
+          "leaf_count": 399
+        },
+        "right_child": {
+          "split_index": 1,
+          "split_feature": 0,
+          "split_gain": 36.74250030517578,
+          "threshold": 0.5102250691730842,
+          "decision_type": "<=",
+          "default_left": true,
+          "missing_type": "NaN",
+          "internal_value": -0.204906,
+          "internal_weight": 601,
+          "internal_count": 601,
+          "left_child": {
+            "leaf_index": 1,
+            "leaf_value": -0.04439151147520909,
+            "leaf_weight": 315,
+            "leaf_count": 315
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": 0.005117411709368601,
+            "leaf_weight": 286,
+            "leaf_count": 286
+          }
+        }
+      }
+    },
+    {
+      "tree_index": 2,
+      "num_leaves": 3,
+      "num_cat": 0,
+      "shrinkage": 0.1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 1,
+        "split_gain": 57.1327018737793,
+        "threshold": 0.668665477622446,
+        "decision_type": "<=",
+        "default_left": true,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "split_index": 1,
+          "split_feature": 1,
+          "split_gain": 40.859100341796875,
+          "threshold": 0.008118820676863816,
+          "decision_type": "<=",
+          "default_left": true,
+          "missing_type": "NaN",
+          "internal_value": -0.162926,
+          "internal_weight": 681,
+          "internal_count": 681,
+          "left_child": {
+            "leaf_index": 0,
+            "leaf_value": -0.15361238490967524,
+            "leaf_weight": 21,
+            "leaf_count": 21
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": -0.01192330846157292,
+            "leaf_weight": 660,
+            "leaf_count": 660
+          }
+        },
+        "right_child": {
+          "leaf_index": 1,
+          "leaf_value": 0.03499044894987518,
+          "leaf_weight": 319,
+          "leaf_count": 319
+        }
+      }
+    },
+    {
+      "tree_index": 3,
+      "num_leaves": 3,
+      "num_cat": 1,
+      "shrinkage": 0.1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 0,
+        "split_gain": 54.77090072631836,
+        "threshold": 0.5201391072644542,
+        "decision_type": "<=",
+        "default_left": true,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "leaf_index": 0,
+          "leaf_value": -0.02141000620783247,
+          "leaf_weight": 543,
+          "leaf_count": 543
+        },
+        "right_child": {
+          "split_index": 1,
+          "split_feature": 2,
+          "split_gain": 27.200700759887695,
+          "threshold": "0||1",
+          "decision_type": "==",
+          "default_left": false,
+          "missing_type": "NaN",
+          "internal_value": 0.255704,
+          "internal_weight": 457,
+          "internal_count": 457,
+          "left_child": {
+            "leaf_index": 1,
+            "leaf_value": -0.004121485787596721,
+            "leaf_weight": 191,
+            "leaf_count": 191
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": 0.04534090904886873,
+            "leaf_weight": 266,
+            "leaf_count": 266
+          }
+        }
+      }
+    },
+    {
+      "tree_index": 4,
+      "num_leaves": 3,
+      "num_cat": 1,
+      "shrinkage": 0.1,
+      "tree_structure": {
+        "split_index": 0,
+        "split_feature": 3,
+        "split_gain": 51.84349822998047,
+        "threshold": "2||3||4",
+        "decision_type": "==",
+        "default_left": false,
+        "missing_type": "NaN",
+        "internal_value": 0,
+        "internal_weight": 0,
+        "internal_count": 1000,
+        "left_child": {
+          "split_index": 1,
+          "split_feature": 1,
+          "split_gain": 39.352699279785156,
+          "threshold": 0.27283279016959255,
+          "decision_type": "<=",
+          "default_left": true,
+          "missing_type": "NaN",
+          "internal_value": 0.188414,
+          "internal_weight": 593,
+          "internal_count": 593,
+          "left_child": {
+            "leaf_index": 0,
+            "leaf_value": -0.01924803254356527,
+            "leaf_weight": 184,
+            "leaf_count": 184
+          },
+          "right_child": {
+            "leaf_index": 2,
+            "leaf_value": 0.03643772842347651,
+            "leaf_weight": 409,
+            "leaf_count": 409
+          }
+        },
+        "right_child": {
+          "leaf_index": 1,
+          "leaf_value": -0.02701711918923075,
+          "leaf_weight": 407,
+          "leaf_count": 407
+        }
+      }
+    }
+  ],
+  "pandas_categorical": [
+    [
+      "a",
+      "b",
+      "c",
+      "d",
+      "e"
+    ],
+    [
+      "i",
+      "j",
+      "k",
+      "l",
+      "m"
+    ]
+  ]
+}
\ No newline at end of file
diff --git a/config-model/src/test/java/com/yahoo/searchdefinition/processing/RankProfileSearchFixture.java b/config-model/src/test/java/com/yahoo/searchdefinition/processing/RankProfileSearchFixture.java
index 08dd5148b29..0cd6674751e 100644
--- a/config-model/src/test/java/com/yahoo/searchdefinition/processing/RankProfileSearchFixture.java
+++ b/config-model/src/test/java/com/yahoo/searchdefinition/processing/RankProfileSearchFixture.java
@@ -15,6 +15,7 @@ import com.yahoo.searchdefinition.parser.ParseException;
 import ai.vespa.rankingexpression.importer.configmodelview.ImportedMlModels;
 import ai.vespa.rankingexpression.importer.onnx.OnnxImporter;
 import ai.vespa.rankingexpression.importer.tensorflow.TensorFlowImporter;
+import ai.vespa.rankingexpression.importer.lightgbm.LightGBMImporter;
 import ai.vespa.rankingexpression.importer.xgboost.XGBoostImporter;
 
 import java.util.HashMap;
@@ -33,6 +34,7 @@ class RankProfileSearchFixture {
 
     private final ImmutableList<MlModelImporter> importers = ImmutableList.of(new TensorFlowImporter(),
                                                                               new OnnxImporter(),
+                                                                              new LightGBMImporter(),
                                                                               new XGBoostImporter());
     private RankProfileRegistry rankProfileRegistry = new RankProfileRegistry();
     private final QueryProfileRegistry queryProfileRegistry;
diff --git a/config-model/src/test/java/com/yahoo/searchdefinition/processing/RankingExpressionWithLightGBMTestCase.java b/config-model/src/test/java/com/yahoo/searchdefinition/processing/RankingExpressionWithLightGBMTestCase.java
new file mode 100644
index 00000000000..79d19371f1c
--- /dev/null
+++ b/config-model/src/test/java/com/yahoo/searchdefinition/processing/RankingExpressionWithLightGBMTestCase.java
@@ -0,0 +1,88 @@
+// Copyright 2020 Oath Inc. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root.
+package com.yahoo.searchdefinition.processing;
+
+import com.yahoo.config.application.api.ApplicationPackage;
+import com.yahoo.io.IOUtils;
+import com.yahoo.path.Path;
+import com.yahoo.searchdefinition.parser.ParseException;
+import org.junit.After;
+import org.junit.Test;
+
+import java.io.IOException;
+
+/**
+ * @author lesters
+ */
+public class RankingExpressionWithLightGBMTestCase {
+
+    private final Path applicationDir = Path.fromString("src/test/integration/lightgbm/");
+
+    private final static String lightGBMExpression =
+            "if (!(numerical_2 >= 0.46643291586559305), 2.1594397038037663, if (categorical_2 in [\"k\", \"l\", \"m\"], 2.235297305276056, 2.1792953471546546)) + if (categorical_1 in [\"d\", \"e\"], 0.03070842919354316, if (!(numerical_1 >= 0.5102250691730842), -0.04439151147520909, 0.005117411709368601)) + if (!(numerical_2 >= 0.668665477622446), if (!(numerical_2 >= 0.008118820676863816), -0.15361238490967524, -0.01192330846157292), 0.03499044894987518) + if (!(numerical_1 >= 0.5201391072644542), -0.02141000620783247, if (categorical_1 in [\"a\", \"b\"], -0.004121485787596721, 0.04534090904886873)) + if (categorical_2 in [\"k\", \"l\", \"m\"], if (!(numerical_2 >= 0.27283279016959255), -0.01924803254356527, 0.03643772842347651), -0.02701711918923075)";
+
+    @After
+    public void removeGeneratedModelFiles() {
+        IOUtils.recursiveDeleteDir(applicationDir.append(ApplicationPackage.MODELS_GENERATED_DIR).toFile());
+    }
+
+    @Test
+    public void testLightGBMReference() {
+        RankProfileSearchFixture search = fixtureWith("lightgbm('regression.json')");
+        search.assertFirstPhaseExpression(lightGBMExpression, "my_profile");
+    }
+
+    @Test
+    public void testNestedLightGBMReference() {
+        RankProfileSearchFixture search = fixtureWith("5 + sum(lightgbm('regression.json'))");
+        search.assertFirstPhaseExpression("5 + reduce(" + lightGBMExpression + ", sum)", "my_profile");
+    }
+
+    @Test
+    public void testImportingFromStoredExpressions() throws IOException {
+        RankProfileSearchFixture search = fixtureWith("lightgbm('regression.json')");
+        search.assertFirstPhaseExpression(lightGBMExpression, "my_profile");
+
+        // At this point the expression is stored - copy application to another location which do not have a models dir
+        Path storedApplicationDirectory = applicationDir.getParentPath().append("copy");
+        try {
+            storedApplicationDirectory.toFile().mkdirs();
+            IOUtils.copyDirectory(applicationDir.append(ApplicationPackage.MODELS_GENERATED_DIR).toFile(),
+                                  storedApplicationDirectory.append(ApplicationPackage.MODELS_GENERATED_DIR).toFile());
+            RankingExpressionWithTensorFlowTestCase.StoringApplicationPackage storedApplication = new RankingExpressionWithTensorFlowTestCase.StoringApplicationPackage(storedApplicationDirectory);
+            RankProfileSearchFixture searchFromStored = fixtureWith("lightgbm('regression.json')");
+            searchFromStored.assertFirstPhaseExpression(lightGBMExpression, "my_profile");
+        }
+        finally {
+            IOUtils.recursiveDeleteDir(storedApplicationDirectory.toFile());
+        }
+    }
+
+    private RankProfileSearchFixture fixtureWith(String firstPhaseExpression) {
+        return fixtureWith(firstPhaseExpression, null, null,
+                new RankingExpressionWithTensorFlowTestCase.StoringApplicationPackage(applicationDir));
+    }
+
+    private RankProfileSearchFixture fixtureWith(String firstPhaseExpression,
+                                                 String constant,
+                                                 String field,
+                                                 RankingExpressionWithTensorFlowTestCase.StoringApplicationPackage application) {
+        try {
+            RankProfileSearchFixture fixture = new RankProfileSearchFixture(
+                    application,
+                    application.getQueryProfiles(),
+                    "  rank-profile my_profile {\n" +
+                            "    first-phase {\n" +
+                            "      expression: " + firstPhaseExpression +
+                            "    }\n" +
+                            "  }",
+                    constant,
+                    field);
+            fixture.compileRankProfile("my_profile", applicationDir.append("models"));
+            return fixture;
+        } catch (ParseException e) {
+            throw new IllegalArgumentException(e);
+        }
+    }
+
+}
+
diff --git a/config-model/src/test/java/com/yahoo/vespa/model/ml/ImportedModelTester.java b/config-model/src/test/java/com/yahoo/vespa/model/ml/ImportedModelTester.java
index ce36ecc4a1c..7a3b76db7f8 100644
--- a/config-model/src/test/java/com/yahoo/vespa/model/ml/ImportedModelTester.java
+++ b/config-model/src/test/java/com/yahoo/vespa/model/ml/ImportedModelTester.java
@@ -1,7 +1,6 @@
 // Copyright 2018 Yahoo Holdings. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root.
 package com.yahoo.vespa.model.ml;
 
-import ai.vespa.rankingexpression.importer.vespa.VespaImporter;
 import com.google.common.collect.ImmutableList;
 import com.yahoo.config.model.ApplicationPackageTester;
 import ai.vespa.rankingexpression.importer.configmodelview.MlModelImporter;
@@ -10,8 +9,10 @@ import com.yahoo.io.GrowableByteBuffer;
 import com.yahoo.io.IOUtils;
 import com.yahoo.path.Path;
 import com.yahoo.searchdefinition.RankingConstant;
+import ai.vespa.rankingexpression.importer.lightgbm.LightGBMImporter;
 import ai.vespa.rankingexpression.importer.onnx.OnnxImporter;
 import ai.vespa.rankingexpression.importer.tensorflow.TensorFlowImporter;
+import ai.vespa.rankingexpression.importer.vespa.VespaImporter;
 import ai.vespa.rankingexpression.importer.xgboost.XGBoostImporter;
 import com.yahoo.tensor.Tensor;
 import com.yahoo.tensor.serialization.TypedBinaryFormat;
@@ -35,6 +36,7 @@ public class ImportedModelTester {
 
     private final ImmutableList<MlModelImporter> importers = ImmutableList.of(new TensorFlowImporter(),
                                                                               new OnnxImporter(),
+                                                                              new LightGBMImporter(),
                                                                               new XGBoostImporter(),
                                                                               new VespaImporter());
 
diff --git a/config-model/src/test/java/com/yahoo/vespa/model/ml/MlModelsTest.java b/config-model/src/test/java/com/yahoo/vespa/model/ml/MlModelsTest.java
index c5c475360a3..ced7243adf5 100644
--- a/config-model/src/test/java/com/yahoo/vespa/model/ml/MlModelsTest.java
+++ b/config-model/src/test/java/com/yahoo/vespa/model/ml/MlModelsTest.java
@@ -45,7 +45,7 @@ public class MlModelsTest {
 
     private void verify(VespaModel model) {
         assertEquals("Global models are created (although not used directly here",
-                     4, model.rankProfileList().getRankProfiles().size());
+                     5, model.rankProfileList().getRankProfiles().size());
 
         RankProfilesConfig.Builder builder = new RankProfilesConfig.Builder();
         model.getSearchClusters().get(0).getConfig(builder);
@@ -71,8 +71,9 @@ public class MlModelsTest {
             "rankingExpression(mnist_softmax_tensorflow).rankingScript: join(reduce(join(rename(rankingExpression(Placeholder), (d0, d1), (d0, d2)), constant(mnist_softmax_saved_layer_Variable_read), f(a,b)(a * b)), sum, d2), constant(mnist_softmax_saved_layer_Variable_1_read), f(a,b)(a + b))\n" +
             "rankingExpression(mnist_softmax_onnx).rankingScript: join(reduce(join(rename(rankingExpression(Placeholder), (d0, d1), (d0, d2)), constant(mnist_softmax_Variable), f(a,b)(a * b)), sum, d2), constant(mnist_softmax_Variable_1), f(a,b)(a + b))\n" +
             "rankingExpression(my_xgboost).rankingScript: if (f29 < -0.1234567, if (!(f56 >= -0.242398), 1.71218, -1.70044), if (f109 < 0.8723473, -1.94071, 1.85965)) + if (!(f60 >= -0.482947), if (f29 < -4.2387498, 0.784718, -0.96853), -6.23624)\n" +
+            "rankingExpression(my_lightgbm).rankingScript: if (!(numerical_2 >= 0.46643291586559305), 2.1594397038037663, if (categorical_2 in [\"k\", \"l\", \"m\"], 2.235297305276056, 2.1792953471546546)) + if (categorical_1 in [\"d\", \"e\"], 0.03070842919354316, if (!(numerical_1 >= 0.5102250691730842), -0.04439151147520909, 0.005117411709368601)) + if (!(numerical_2 >= 0.668665477622446), if (!(numerical_2 >= 0.008118820676863816), -0.15361238490967524, -0.01192330846157292), 0.03499044894987518) + if (!(numerical_1 >= 0.5201391072644542), -0.02141000620783247, if (categorical_1 in [\"a\", \"b\"], -0.004121485787596721, 0.04534090904886873)) + if (categorical_2 in [\"k\", \"l\", \"m\"], if (!(numerical_2 >= 0.27283279016959255), -0.01924803254356527, 0.03643772842347651), -0.02701711918923075)\n" +
             "vespa.rank.firstphase: rankingExpression(firstphase)\n" +
-            "rankingExpression(firstphase).rankingScript: rankingExpression(mnist_tensorflow) + rankingExpression(mnist_softmax_tensorflow) + rankingExpression(mnist_softmax_onnx) + rankingExpression(my_xgboost)\n" +
+            "rankingExpression(firstphase).rankingScript: rankingExpression(mnist_tensorflow) + rankingExpression(mnist_softmax_tensorflow) + rankingExpression(mnist_softmax_onnx) + rankingExpression(my_xgboost) + rankingExpression(my_lightgbm)\n" +
             "vespa.type.attribute.argument: tensor<float>(d0[],d1[784])\n";
 
 }
diff --git a/config-model/src/test/java/com/yahoo/vespa/model/ml/ModelEvaluationTest.java b/config-model/src/test/java/com/yahoo/vespa/model/ml/ModelEvaluationTest.java
index 3d4ac7f2eeb..2d3ddc33afb 100644
--- a/config-model/src/test/java/com/yahoo/vespa/model/ml/ModelEvaluationTest.java
+++ b/config-model/src/test/java/com/yahoo/vespa/model/ml/ModelEvaluationTest.java
@@ -96,9 +96,10 @@ public class ModelEvaluationTest {
         cluster.getConfig(cb);
         RankingConstantsConfig constantsConfig = new RankingConstantsConfig(cb);
 
-        assertEquals(4, config.rankprofile().size());
+        assertEquals(5, config.rankprofile().size());
         Set<String> modelNames = config.rankprofile().stream().map(v -> v.name()).collect(Collectors.toSet());
         assertTrue(modelNames.contains("xgboost_2_2"));
+        assertTrue(modelNames.contains("lightgbm_regression"));
         assertTrue(modelNames.contains("mnist_saved"));
         assertTrue(modelNames.contains("mnist_softmax"));
         assertTrue(modelNames.contains("mnist_softmax_saved"));
@@ -112,13 +113,18 @@ public class ModelEvaluationTest {
         ModelsEvaluator evaluator = new ModelsEvaluator(new ToleratingMissingConstantFilesRankProfilesConfigImporter(MockFileAcquirer.returnFile(null))
                                                                 .importFrom(config, constantsConfig));
 
-        assertEquals(4, evaluator.models().size());
+        assertEquals(5, evaluator.models().size());
 
         Model xgboost = evaluator.models().get("xgboost_2_2");
         assertNotNull(xgboost);
         assertNotNull(xgboost.evaluatorOf());
         assertNotNull(xgboost.evaluatorOf("xgboost_2_2"));
 
+        Model lightgbm = evaluator.models().get("lightgbm_regression");
+        assertNotNull(lightgbm);
+        assertNotNull(lightgbm.evaluatorOf());
+        assertNotNull(lightgbm.evaluatorOf("lightgbm_regression"));
+
         Model tensorflow_mnist = evaluator.models().get("mnist_saved");
         assertNotNull(tensorflow_mnist);
         assertEquals(1, tensorflow_mnist.functions().size());