Adding OneHotHashEncoding Test (#5098)

Author: Lynx1820

test/BaselineOutput/Common/Onnx/BinaryClassification/BreastCancer/ExcludeVariablesInOnnxConversion.txt

@@ -447,14 +447,24 @@
           }
         ]
       },
+      {
+        "input": [
+          "F3"
+        ],
+        "output": [
+          "F3.output"
+        ],
+        "name": "Identity",
+        "opType": "Identity"
+      },
       {
         "input": [
           "PredictedLabel"
         ],
         "output": [
           "PredictedLabel.output"
         ],
-        "name": "Identity",
+        "name": "Identity0",
         "opType": "Identity"
       },
       {
@@ -464,7 +474,7 @@
         "output": [
           "Score.output"
         ],
-        "name": "Identity0",
+        "name": "Identity1",
         "opType": "Identity"
       },
       {
@@ -474,7 +484,7 @@
         "output": [
           "Probability.output"
         ],
-        "name": "Identity1",
+        "name": "Identity2",
         "opType": "Identity"
       }
     ],
@@ -531,9 +541,45 @@
             }
           }
         }
+      },
+      {
+        "name": "F3",
+        "type": {
+          "tensorType": {
+            "elemType": 8,
+            "shape": {
+              "dim": [
+                {
+                  "dimValue": "-1"
+                },
+                {
+                  "dimValue": "5"
+                }
+              ]
+            }
+          }
+        }
       }
     ],
     "output": [
+      {
+        "name": "F3.output",
+        "type": {
+          "tensorType": {
+            "elemType": 8,
+            "shape": {
+              "dim": [
+                {
+                  "dimValue": "-1"
+                },
+                {
+                  "dimValue": "5"
+                }
+              ]
+            }
+          }
+        }
+      },
       {
         "name": "PredictedLabel.output",
         "type": {
@@ -806,6 +852,24 @@
           }
         }
       },
+      {
+        "name": "F3.output",
+        "type": {
+          "tensorType": {
+            "elemType": 8,
+            "shape": {
+              "dim": [
+                {
+                  "dimValue": "-1"
+                },
+                {
+                  "dimValue": "5"
+                }
+              ]
+            }
+          }
+        }
+      },
       {
         "name": "PredictedLabel.output",
         "type": {

test/Microsoft.ML.Tests/OnnxConversionTest.cs

@@ -9,7 +9,6 @@
 using System.Runtime.InteropServices;
 using System.Text.RegularExpressions;
 using Google.Protobuf;
-using Google.Protobuf.WellKnownTypes;
 using Microsoft.ML.Data;
 using Microsoft.ML.EntryPoints;
 using Microsoft.ML.Model.OnnxConverter;
@@ -125,6 +124,9 @@ private class BreastCancerCatFeatureExample
 
             [LoadColumn(2)]
             public string F2;
+
+            [LoadColumn(3, 7), VectorType(6)]
+            public string[] F3;
         }
 
         private class BreastCancerMulticlassExample
@@ -1162,6 +1164,37 @@ public void PcaOnnxConversionTest()
             Done();
         }
 
+        [Fact]
+        public void OneHotHashEncodingOnnxConversionTest()
+        {
+            var mlContext = new MLContext();
+            string dataPath = GetDataPath("breast-cancer.txt");
+
+            var dataView = ML.Data.LoadFromTextFile<BreastCancerCatFeatureExample>(dataPath);
+            var pipe = ML.Transforms.Categorical.OneHotHashEncoding(new[]{
+                    new OneHotHashEncodingEstimator.ColumnOptions("Output", "F3", useOrderedHashing:false),
+                });
+            var model = pipe.Fit(dataView);
+            var transformedData = model.Transform(dataView);
+            var onnxModel = mlContext.Model.ConvertToOnnxProtobuf(model, dataView);
+
+            var onnxFileName = "OneHotHashEncoding.onnx";
+            var onnxModelPath = GetOutputPath(onnxFileName);
+            SaveOnnxModel(onnxModel, onnxModelPath, null);
+
+            if (IsOnnxRuntimeSupported())
+            {
+                // Evaluate the saved ONNX model using the data used to train the ML.NET pipeline.
+                string[] inputNames = onnxModel.Graph.Input.Select(valueInfoProto => valueInfoProto.Name).ToArray();
+                string[] outputNames = onnxModel.Graph.Output.Select(valueInfoProto => valueInfoProto.Name).ToArray();
+                var onnxEstimator = mlContext.Transforms.ApplyOnnxModel(outputNames, inputNames, onnxModelPath);
+                var onnxTransformer = onnxEstimator.Fit(dataView);
+                var onnxResult = onnxTransformer.Transform(dataView);
+                CompareSelectedColumns<float>("Output", "Output", transformedData, onnxResult);
+            }
+            Done();
+        }
+
         [Theory]
         [CombinatorialData]
         // Due to lack of Onnxruntime support, long/ulong, double, floats, and OrderedHashing are not supported.