Addressed reviewers' comments.

Author: zeahmed

src/Microsoft.ML.TensorFlow/TensorFlowModel.cs

@@ -80,14 +80,54 @@ public TensorFlowEstimator ScoreTensorFlowModel(string[] outputColumnNames, stri
             => new TensorFlowEstimator(_env, outputColumnNames, inputColumnNames, ModelPath);
 
         /// <summary>
-        /// Create the <see cref="TensorFlowEstimator"/> for scoring or retraining using the tensorflow model.
+        /// Retrain the TensorFlow model on new data.
         /// The model is not loaded again instead the information contained in <see cref="TensorFlowModel"/> class is reused
         /// (c.f. <see cref="TensorFlowModel.ModelPath"/> and <see cref="TensorFlowModel.Session"/>).
         /// </summary>
-        /// <param name="options">The <see cref="TensorFlowEstimator.Options"/> specifying the inputs and the settings of the <see cref="TensorFlowEstimator"/>.</param>
-        public TensorFlowEstimator CreateTensorFlowEstimator(TensorFlowEstimator.Options options)
+        /// <param name="inputColumnNames"> The names of the model inputs.</param>
+        /// <param name="outputColumnNames">The names of the requested model outputs.</param>
+        /// <param name="labelColumnName">Name of the label column.</param>
+        /// <param name="tensorFlowLabel">Name of the node in TensorFlow graph that is used as label during training in TensorFlow.
+        /// The value of <paramref name="labelColumnName"/> from <see cref="IDataView"/> is fed to this node.</param>
+        /// <param name="optimizationOperation">The name of the optimization operation in the TensorFlow graph.</param>
+        /// <param name="epoch">Number of training iterations.</param>
+        /// <param name="batchSize">Number of samples to use for mini-batch training.</param>
+        /// <param name="lossOperation">The name of the operation in the TensorFlow graph to compute training loss (Optional).</param>
+        /// <param name="metricOperation">The name of the operation in the TensorFlow graph to compute performance metric during training (Optional).</param>
+        /// <param name="learningRateOperation">The name of the operation in the TensorFlow graph which sets optimizer learning rate (Optional).</param>
+        /// <param name="learningRate">Learning rate to use during optimization (Optional).</param>
+        /// <remarks>
+        /// The support for retraining is experimental.
+        /// </remarks>
+        public TensorFlowEstimator RetrainTensorFlowModel(
+            string[] outputColumnNames,
+            string[] inputColumnNames,
+            string labelColumnName,
+            string tensorFlowLabel,
+            string optimizationOperation,
+            int epoch = 10,
+            int batchSize = 20,
+            string lossOperation= null,
+            string metricOperation = null,
+            string learningRateOperation = null,
+            float learningRate = 0.01f)
         {
-            options.ModelLocation = ModelPath;
+            var options = new TensorFlowEstimator.Options()
+            {
+                ModelLocation = ModelPath,
+                InputColumns = inputColumnNames,
+                OutputColumns = outputColumnNames,
+                LabelColumn = labelColumnName,
+                TensorFlowLabel = tensorFlowLabel,
+                OptimizationOperation = optimizationOperation,
+                LossOperation = lossOperation,
+                MetricOperation = metricOperation,
+                Epoch = epoch,
+                LearningRateOperation = learningRateOperation,
+                LearningRate = learningRate,
+                BatchSize = batchSize,
+                ReTrain = true
+            };
             return new TensorFlowEstimator(_env, options, this);
         }
     }

src/Microsoft.ML.TensorFlow/TensorflowCatalog.cs

@@ -14,7 +14,7 @@ public static class TensorflowCatalog
     {
         /// <summary>
         /// Load TensorFlow model into memory. This is the convenience method that allows the model to be loaded once and subsequently use it for querying schema and creation of
-        /// <see cref="TensorFlowEstimator"/> using <see cref="TensorFlowModel.CreateTensorFlowEstimator(TensorFlowEstimator.Options)"/>.
+        /// <see cref="TensorFlowEstimator"/> using <see cref="TensorFlowModel.ScoreTensorFlowModel(string, string)"/>.
         /// </summary>
         /// <param name="catalog">The transform's catalog.</param>
         /// <param name="modelLocation">Location of the TensorFlow model.</param>

test/Microsoft.ML.Tests/ScenariosWithDirectInstantiation/TensorflowTests.cs

@@ -544,20 +544,17 @@ public void TensorFlowTransformMNISTLRTrainingTest()
 
                 var pipe = mlContext.Transforms.Categorical.OneHotEncoding("OneHotLabel", "Label")
                     .Append(mlContext.Transforms.Normalize(new NormalizingEstimator.MinMaxColumnOptions("Features", "Placeholder")))
-                    .Append(mlContext.Model.LoadTensorFlowModel(model_location).CreateTensorFlowEstimator(new TensorFlowEstimator.Options()
-                    {
-                        InputColumns = new[] { "Features" },
-                        OutputColumns = new[] { "Prediction", "b" },
-                        LabelColumn = "OneHotLabel",
-                        TensorFlowLabel = "Label",
-                        OptimizationOperation = "SGDOptimizer",
-                        LossOperation = "Loss",
-                        Epoch = 10,
-                        LearningRateOperation = "SGDOptimizer/learning_rate",
-                        LearningRate = 0.001f,
-                        BatchSize = 20,
-                        ReTrain = true
-                    }))
+                    .Append(mlContext.Model.LoadTensorFlowModel(model_location).RetrainTensorFlowModel(
+                        inputColumnNames: new[] { "Features" },
+                        outputColumnNames: new[] { "Prediction", "b" },
+                        labelColumnName: "OneHotLabel",
+                        tensorFlowLabel: "Label",
+                        optimizationOperation: "SGDOptimizer",
+                        lossOperation: "Loss",
+                        epoch: 10,
+                        learningRateOperation: "SGDOptimizer/learning_rate",
+                        learningRate: 0.001f,
+                        batchSize: 20))
                     .Append(mlContext.Transforms.Concatenate("Features", "Prediction"))
                     .Append(mlContext.Transforms.Conversion.MapValueToKey("KeyLabel", "Label", maxNumKeys: 10))
                     .Append(mlContext.MulticlassClassification.Trainers.LightGbm("KeyLabel", "Features"));
@@ -661,21 +658,18 @@ private void ExecuteTFTransformMNISTConvTrainingTest(bool shuffle, int? shuffleS
                 }
 
                 var pipe = mlContext.Transforms.CopyColumns(("Features", "Placeholder"))
-                    .Append(mlContext.Model.LoadTensorFlowModel(modelLocation).CreateTensorFlowEstimator(new TensorFlowEstimator.Options()
-                    {
-                        InputColumns = new[] { "Features" },
-                        OutputColumns = new[] { "Prediction" },
-                        LabelColumn = "TfLabel",
-                        TensorFlowLabel = "Label",
-                        OptimizationOperation = "MomentumOp",
-                        LossOperation = "Loss",
-                        MetricOperation = "Accuracy",
-                        Epoch = 10,
-                        LearningRateOperation = "learning_rate",
-                        LearningRate = 0.01f,
-                        BatchSize = 20,
-                        ReTrain = true
-                    }))
+                    .Append(mlContext.Model.LoadTensorFlowModel(modelLocation).RetrainTensorFlowModel(
+                        inputColumnNames: new[] { "Features" },
+                        outputColumnNames: new[] { "Prediction" },
+                        labelColumnName: "TfLabel",
+                        tensorFlowLabel: "Label",
+                        optimizationOperation: "MomentumOp",
+                        lossOperation: "Loss",
+                        metricOperation: "Accuracy",
+                        epoch: 10,
+                        learningRateOperation: "learning_rate",
+                        learningRate: 0.01f,
+                        batchSize: 20))
                     .Append(mlContext.Transforms.Concatenate("Features", "Prediction"))
                     .AppendCacheCheckpoint(mlContext)
                     .Append(mlContext.MulticlassClassification.Trainers.LightGbm(new LightGBM.Options()