Ensure ONNX export is compatible with Windows RS5

src/Microsoft.ML.Data/Prediction/Calibrator.cs

@@ -1437,13 +1437,13 @@ public bool SaveAsOnnx(OnnxContext ctx, string[] scoreProbablityColumnNames, str
             string opType = "Affine";
             string linearOutput = ctx.AddIntermediateVariable(null, "linearOutput", true);
             var node = ctx.CreateNode(opType, new[] { scoreProbablityColumnNames[0] },
-                new[] { linearOutput }, ctx.GetNodeName(opType), "ai.onnx");
+                new[] { linearOutput }, ctx.GetNodeName(opType), "");
             node.AddAttribute("alpha", ParamA * -1);
             node.AddAttribute("beta", -0.0000001);
 
             opType = "Sigmoid";
             node = ctx.CreateNode(opType, new[] { linearOutput },
-                new[] { scoreProbablityColumnNames[1] }, ctx.GetNodeName(opType), "ai.onnx");
+                new[] { scoreProbablityColumnNames[1] }, ctx.GetNodeName(opType), "");
 
             return true;
         }

src/Microsoft.ML.Data/Transforms/ConcatTransform.cs

@@ -723,7 +723,6 @@ public void SaveAsOnnx(OnnxContext ctx)
                 var node = ctx.CreateNode(opType, inputList.Select(t => t.Key),
                     new[] { ctx.AddIntermediateVariable(outColType, outName) }, ctx.GetNodeName(opType));
 
-                node.AddAttribute("inputList", inputList.Select(x => x.Key));
                 node.AddAttribute("inputdimensions", inputList.Select(x => x.Value));
             }
         }

src/Microsoft.ML.Data/Transforms/TermTransform.cs

@@ -719,7 +719,10 @@ protected override bool SaveAsOnnxCore(OnnxContext ctx, int iinfo, ColInfo info,
             var node = ctx.CreateNode(opType, srcVariableName, dstVariableName, ctx.GetNodeName(opType));
             node.AddAttribute("classes_strings", terms.DenseValues());
             node.AddAttribute("default_int64", -1);
-            node.AddAttribute("default_string", DvText.Empty);
+            //default_string needs to be an empty string but there is a BUG in Lotus that 
+            //throws a validation error when default_string is empty. As a work around, set
+            //default_string to a space.
+            node.AddAttribute("default_string", " ");
             return true;
         }
 

src/Microsoft.ML.Onnx/OnnxUtils.cs

@@ -252,10 +252,10 @@ public static ModelProto MakeModel(List<NodeProto> nodes, string producerName, s
             model.Domain = domain;
             model.ProducerName = producerName;
             model.ProducerVersion = producerVersion;
-            model.IrVersion = (long)UniversalModelFormat.Onnx.Version.IrVersion;
+            model.IrVersion = (long)Version.IrVersion;
             model.ModelVersion = modelVersion;
             model.OpsetImport.Add(new OperatorSetIdProto() { Domain = "ai.onnx.ml", Version = 1 });
-            model.OpsetImport.Add(new OperatorSetIdProto() { Domain = "ai.onnx", Version = 6 });
+            model.OpsetImport.Add(new OperatorSetIdProto() { Domain = "", Version = 7 });
             model.Graph = new GraphProto();
             var graph = model.Graph;
             graph.Node.Add(nodes);

src/Microsoft.ML.StandardLearners/Standard/LinearPredictor.cs

@@ -238,10 +238,10 @@ public bool SaveAsOnnx(OnnxContext ctx, string[] outputs, string featureColumn)
             string opType = "LinearRegressor";
             var node = ctx.CreateNode(opType, new[] { featureColumn }, outputs, ctx.GetNodeName(opType));
             // Selection of logit or probit output transform. enum {'NONE', 'LOGIT', 'PROBIT}
-            node.AddAttribute("post_transform", 0);
+            node.AddAttribute("post_transform", "NONE");
             node.AddAttribute("targets", 1);
             node.AddAttribute("coefficients", Weight.DenseValues());
-            node.AddAttribute("intercepts", Bias);
+            node.AddAttribute("intercepts", new float[] { Bias });
             return true;
         }
 

src/Microsoft.ML.StandardLearners/Standard/LogisticRegression/MulticlassLogisticRegression.cs

@@ -845,12 +845,12 @@ public bool SaveAsOnnx(OnnxContext ctx, string[] outputs, string featureColumn)
 
             string opType = "LinearClassifier";
             var node = ctx.CreateNode(opType, new[] { featureColumn }, outputs, ctx.GetNodeName(opType));
-            // Selection of logit or probit output transform. enum {'NONE', 'LOGIT', 'PROBIT}
-            node.AddAttribute("post_transform", 0);
+            // Selection of logit or probit output transform. enum {'NONE', 'SOFTMAX', 'LOGISTIC', 'SOFTMAX_ZERO', 'PROBIT}
+            node.AddAttribute("post_transform", "NONE");
             node.AddAttribute("multi_class", true);
             node.AddAttribute("coefficients", _weights.SelectMany(w => w.DenseValues()));
             node.AddAttribute("intercepts", _biases);
-            node.AddAttribute("classlabels_strings", _labelNames);
+            node.AddAttribute("classlabels_ints", Enumerable.Range(0, _numClasses).Select(x => (long)x));
             return true;
         }
 

src/Microsoft.ML.Transforms/NAReplaceTransform.cs

@@ -633,13 +633,13 @@ protected override bool SaveAsOnnxCore(OnnxContext ctx, int iinfo, ColInfo info,
             node.AddAttribute("replaced_value_float", Single.NaN);
 
             if (!Infos[iinfo].TypeSrc.IsVector)
-                node.AddAttribute("imputed_value_float", Enumerable.Repeat((float)_repValues[iinfo], 1));
+                node.AddAttribute("imputed_value_floats", Enumerable.Repeat((float)_repValues[iinfo], 1));
             else
             {
                 if (_repIsDefault[iinfo] != null)
                     node.AddAttribute("imputed_value_floats", (float[])_repValues[iinfo]);
                 else
-                    node.AddAttribute("imputed_value_float", Enumerable.Repeat((float)_repValues[iinfo], 1));
+                    node.AddAttribute("imputed_value_floats", Enumerable.Repeat((float)_repValues[iinfo], 1));
             }
 
             return true;