merge

TensorFlow.NET.sln

@@ -15,8 +15,6 @@ Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "TensorFlowNET.Hub", "src\Te
 EndProject
 Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "TensorFlowNET.Datasets", "src\TensorFlowNET.Datasets\TensorFlowNET.Datasets.csproj", "{494D6CAD-2C0D-4C0B-90E2-B097DB039383}"
 EndProject
-Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "NumSharp.Core", "..\NumSharp\src\NumSharp.Core\NumSharp.Core.csproj", "{9249BCC4-3FEB-4EF5-8AB9-789FFE4040B4}"
-EndProject
 Global
 	GlobalSection(SolutionConfigurationPlatforms) = preSolution
 		Debug|Any CPU = Debug|Any CPU
@@ -99,18 +97,6 @@ Global
 		{494D6CAD-2C0D-4C0B-90E2-B097DB039383}.Release|Any CPU.Build.0 = Release|Any CPU
 		{494D6CAD-2C0D-4C0B-90E2-B097DB039383}.Release|x64.ActiveCfg = Release|Any CPU
 		{494D6CAD-2C0D-4C0B-90E2-B097DB039383}.Release|x64.Build.0 = Release|Any CPU
-		{9249BCC4-3FEB-4EF5-8AB9-789FFE4040B4}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
-		{9249BCC4-3FEB-4EF5-8AB9-789FFE4040B4}.Debug|Any CPU.Build.0 = Debug|Any CPU
-		{9249BCC4-3FEB-4EF5-8AB9-789FFE4040B4}.Debug|x64.ActiveCfg = Debug|x64
-		{9249BCC4-3FEB-4EF5-8AB9-789FFE4040B4}.Debug|x64.Build.0 = Debug|x64
-		{9249BCC4-3FEB-4EF5-8AB9-789FFE4040B4}.Publish|Any CPU.ActiveCfg = Publish|Any CPU
-		{9249BCC4-3FEB-4EF5-8AB9-789FFE4040B4}.Publish|Any CPU.Build.0 = Publish|Any CPU
-		{9249BCC4-3FEB-4EF5-8AB9-789FFE4040B4}.Publish|x64.ActiveCfg = Publish|x64
-		{9249BCC4-3FEB-4EF5-8AB9-789FFE4040B4}.Publish|x64.Build.0 = Publish|x64
-		{9249BCC4-3FEB-4EF5-8AB9-789FFE4040B4}.Release|Any CPU.ActiveCfg = Release|Any CPU
-		{9249BCC4-3FEB-4EF5-8AB9-789FFE4040B4}.Release|Any CPU.Build.0 = Release|Any CPU
-		{9249BCC4-3FEB-4EF5-8AB9-789FFE4040B4}.Release|x64.ActiveCfg = Release|x64
-		{9249BCC4-3FEB-4EF5-8AB9-789FFE4040B4}.Release|x64.Build.0 = Release|x64
 	EndGlobalSection
 	GlobalSection(SolutionProperties) = preSolution
 		HideSolutionNode = FALSE

src/SciSharp.TensorFlow.Redist/README.md

@@ -26,7 +26,7 @@ Related merged [commits](https://github.com/SciSharp/TensorFlow.NET/commit/854a5
 
 On Windows, the tar command does not support extracting archives with symlinks. So when `dotnet pack` runs on Windows it will only package the Windows binaries.
 
-1. Run `dotnet pack SciSharp.TensorFlow.Redist-CPU.nupkgproj` under `src/SciSharp.TensorFlow.Redist` directory in Linux.
-2. Run `dotnet nuget push SciSharp.TensorFlow.Redist.1.14.0.nupkg -k APIKEY -s https://api.nuget.org/v3/index.json`
+1. Run `dotnet pack SciSharp.TensorFlow.Redist.nupkgproj` under `src/SciSharp.TensorFlow.Redist` directory in Linux.
+2. Run `dotnet nuget push SciSharp.TensorFlow.Redist.1.15.0.nupkg -k APIKEY -s https://api.nuget.org/v3/index.json`
 
 

src/SciSharp.TensorFlow.Redist/SciSharp.TensorFlow.Redist-CPU.nupkgproj → src/SciSharp.TensorFlow.Redist/SciSharp.TensorFlow.Redist.nupkgproj

@@ -7,7 +7,7 @@
 
     <TargetArchitecture Condition="'$(TargetArchitecture)' == ''">x64</TargetArchitecture>
     <TargetFramework>netstandard2.0</TargetFramework>
-    <TensorFlowVersion>1.14.0</TensorFlowVersion>
+    <TensorFlowVersion>1.15.0</TensorFlowVersion>
     <TensorFlowMajorVersion>1</TensorFlowMajorVersion>
 
     <PackageAssetsPath>$(BinDir)packages\</PackageAssetsPath>
@@ -38,7 +38,7 @@
     <TensorFlowConfig Include="windows"
                       FileExtension=".zip"
                       FilesFromArchive="lib\tensorflow.dll;
-                                        include\tensorflow\c\LICENSE"
+                                        LICENSE"
                       Runtime="win-x64"/>
 
     <TensorFlowConfig Condition="'$(OS)' != 'Windows_NT'"
@@ -46,14 +46,14 @@
                       FileExtension=".tar.gz"
                       FilesFromArchive="lib\libtensorflow.so;
                                         lib\libtensorflow_framework.so.$(TensorFlowMajorVersion);
-                                        include\tensorflow\c\LICENSE"
+                                        LICENSE"
                       Runtime="linux-x64" />
 
     <TensorFlowConfig Condition="'$(OS)' != 'Windows_NT'"
                       Include="darwin" FileExtension=".tar.gz"
                       FilesFromArchive="lib\libtensorflow.dylib;
                                         lib\libtensorflow_framework.$(TensorFlowMajorVersion).dylib;
-                                        include\tensorflow\c\LICENSE"
+                                        LICENSE"
                       Runtime="osx-x64" />
 
     <AdditionalDownloadFile Include="https://raw.githubusercontent.com/tensorflow/tensorflow/master/LICENSE"

src/TensorFlowNET.Core/APIs/tf.train.cs

@@ -34,6 +34,9 @@ namespace Tensorflow
             public Optimizer GradientDescentOptimizer(float learning_rate) 
                 => new GradientDescentOptimizer(learning_rate);
 
+            public Optimizer GradientDescentOptimizer(Tensor learning_rate)
+                => new GradientDescentOptimizer(learning_rate);
+
             public Optimizer AdamOptimizer(float learning_rate, string name = "Adam") 
                 => new AdamOptimizer(learning_rate, name: name);
 

src/TensorFlowNET.Core/Binding.Util.cs

@@ -129,6 +129,7 @@ namespace Tensorflow
             }
         }
 
+        [DebuggerStepThrough]
         [DebuggerNonUserCode()] // with "Just My Code" enabled this lets the debugger break at the origin of the exception
         public static TOut tf_with<TIn, TOut>(TIn py, Func<TIn, TOut> action) where TIn : IObjectLife
         {

src/TensorFlowNET.Core/Framework/meta_graph.cs

@@ -134,7 +134,7 @@ namespace Tensorflow
                                         }
                                         break;
                                     default:
-                                        Console.WriteLine("import_scoped_meta_graph_with_return_elements");
+                                        Console.WriteLine($"import_scoped_meta_graph_with_return_elements {col.Key}");
                                         continue;
                                 }
                             }

src/TensorFlowNET.Core/Framework/smart_module.cs

@@ -15,6 +15,7 @@
 ******************************************************************************/
 
 using System;
+using static Tensorflow.Binding;
 
 namespace Tensorflow.Framework
 {

src/TensorFlowNET.Core/Graphs/Graph.cs

@@ -75,7 +75,7 @@ namespace Tensorflow
     ///     then create a TensorFlow session to run parts of the graph across a set of local and remote devices.
     /// </summary>
     /// <remarks>https://www.tensorflow.org/guide/graphs <br></br>https://www.tensorflow.org/api_docs/python/tf/Graph</remarks>
-    public partial class Graph : DisposableObject, IEnumerable<Operation>
+    public partial class Graph : DisposableObject//, IEnumerable<Operation>
     {
         private Dictionary<int, ITensorOrOperation> _nodes_by_id;
         public Dictionary<string, ITensorOrOperation> _nodes_by_name;
@@ -257,17 +257,17 @@ namespace Tensorflow
             if (inputs == null)
                 inputs = new Tensor[0];
 
-            foreach ((int idx, Tensor a) in enumerate(inputs))
-            {
-
-            }
-
-            if (String.IsNullOrEmpty(name))
+            if (string.IsNullOrEmpty(name))
                 name = op_type;
             // If a names ends with a '/' it is a "name scope" and we use it as-is,
             // after removing the trailing '/'.
             name = name.EndsWith("/") ? ops.name_from_scope_name(name) : unique_name(name);
             var node_def = ops._NodeDef(op_type, name, device: "", attrs: attrs);
+
+            if (name.Contains("define_loss/bigger_box_loss/mul_13"))
+            {
+
+            }
 
             var input_ops = inputs.Select(x => x.op).ToArray();
             var control_inputs = _control_dependencies_for_inputs(input_ops);
@@ -526,14 +526,14 @@ namespace Tensorflow
             return debugString;*/
         }
 
-        private IEnumerable<Operation> GetEnumerable()
+        /*private IEnumerable<Operation> GetEnumerable()
             => c_api_util.tf_operations(this);
 
         IEnumerator<Operation> IEnumerable<Operation>.GetEnumerator()
             => GetEnumerable().GetEnumerator();
 
         IEnumerator IEnumerable.GetEnumerator() 
-            => throw new NotImplementedException();
+            => throw new NotImplementedException();*/
 
         public static implicit operator IntPtr(Graph graph)
         {

src/TensorFlowNET.Core/Keras/Engine/Sequential.cs

@@ -56,9 +56,9 @@ namespace Tensorflow.Keras.Engine
                     {
                         // Instantiate an input layer.
                         var x = keras.layers.Input(
-                          batch_shape: batch_shape,
-                          dtype: dtype,
-                          name: layer.name + "_input");
+                              batch_shape: batch_shape,
+                              dtype: dtype,
+                              name: layer.name + "_input");
 
                         // This will build the current layer
                         // and create the node connecting the current layer
@@ -71,7 +71,7 @@ namespace Tensorflow.Keras.Engine
                 if (set_inputs)
                 {
                     // If an input layer (placeholder) is available.
-                    // outputs = layer._inbound_nodes;
+                    // outputs = layer.inbound_nodes;
                 }
 
             }

src/TensorFlowNET.Core/Keras/Layers/Layer.cs

@@ -106,6 +106,7 @@ namespace Tensorflow.Keras.Layers
             VariableScope scope = null)
         {
             var input_list = inputs;
+            var input = inputs[0];
             Tensor outputs = null;
 
             // We will attempt to build a TF graph if & only if all inputs are symbolic.
@@ -139,6 +140,7 @@ namespace Tensorflow.Keras.Layers
                     _maybe_build(inputs[0]);
 
                     outputs = call(inputs[0], training: training);
+                    (input, outputs) = _set_connectivity_metadata_(input, outputs);
                     _handle_activity_regularization(inputs[0], outputs);
                     _set_mask_metadata(inputs[0], outputs, null);
                 });
@@ -147,6 +149,12 @@ namespace Tensorflow.Keras.Layers
             return outputs;
         }
 
+        private (Tensor, Tensor) _set_connectivity_metadata_(Tensor inputs, Tensor outputs)
+        {
+            //_add_inbound_node(input_tensors: inputs, output_tensors: outputs);
+            return (inputs, outputs);
+        }
+
         private void _handle_activity_regularization(Tensor inputs, Tensor outputs)
         {
             //if(_activity_regularizer != null)
@@ -224,7 +232,7 @@ namespace Tensorflow.Keras.Layers
                 overwrite: true,
                 initializer: initializer,
                 trainable: trainable.Value);
-            backend.track_variable(variable);
+            //backend.track_variable(variable);
             _trainable_weights.Add(variable);
 
             return variable;

src/TensorFlowNET.Core/Operations/NnOps/gen_nn_ops.cs

@@ -14,6 +14,8 @@
    limitations under the License.
 ******************************************************************************/
 
+using static Tensorflow.Binding;
+
 namespace Tensorflow.Operations
 {
     public class gen_nn_ops

src/TensorFlowNET.Core/Operations/Operation.Control.cs

@@ -54,10 +54,6 @@ namespace Tensorflow
 
         public void _set_control_flow_context(ControlFlowContext ctx)
         {
-            if(name == "define_loss/conv_sobj_branch/batch_normalization/cond/FusedBatchNorm_1")
-            {
-
-            }
             _control_flow_context = ctx;
         }
 

src/TensorFlowNET.Core/Operations/Operation.Input.cs

@@ -14,6 +14,7 @@
    limitations under the License.
 ******************************************************************************/
 
+using Newtonsoft.Json;
 using System;
 using System.Linq;
 using System.Runtime.InteropServices;
@@ -37,7 +38,9 @@ namespace Tensorflow
             }
             return num;
         }
-
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public int NumInputs => c_api.TF_OperationNumInputs(_handle);
         private TF_DataType[] _input_types => _inputs._inputs.Select(x => x.dtype).ToArray();
 

src/TensorFlowNET.Core/Operations/Operation.Output.cs

@@ -14,6 +14,7 @@
    limitations under the License.
 ******************************************************************************/
 
+using Newtonsoft.Json;
 using System;
 using System.Linq;
 using System.Runtime.InteropServices;
@@ -23,6 +24,9 @@ namespace Tensorflow
 {
     public partial class Operation
     {
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public int NumOutputs => c_api.TF_OperationNumOutputs(_handle);
         public TF_DataType OutputType(int index) => c_api.TF_OperationOutputType(_tf_output(index));
 
@@ -40,7 +44,9 @@ namespace Tensorflow
 
         private Tensor[] _outputs;
         public Tensor[] outputs => _outputs;
-
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public Tensor output => _outputs.FirstOrDefault();
 
         public int NumControlOutputs => c_api.TF_OperationNumControlOutputs(_handle);

src/TensorFlowNET.Core/Operations/Operation.cs

@@ -15,6 +15,9 @@
 ******************************************************************************/
 
 using Google.Protobuf.Collections;
+#if SERIALIZABLE
+using Newtonsoft.Json;
+#endif
 using System;
 using System.Collections.Generic;
 using System.IO;
@@ -43,20 +46,37 @@ namespace Tensorflow
     /// </summary>
     public partial class Operation : ITensorOrOperation
     {
-        private readonly IntPtr _handle; // _c_op in python
-        private readonly Graph _graph;
-        private NodeDef _node_def;
+        private readonly IntPtr _handle; // _c_op in python
 
-        public string type => OpType;
-        public Graph graph => _graph;
-        public int _id => _id_value;
-        public int _id_value;
+        private readonly Graph _graph;
+        private NodeDef _node_def;
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
+        public string type => OpType;
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
+        public Graph graph => _graph;
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
+        public int _id => _id_value;
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
+        public int _id_value;
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public Operation op => this;
         public TF_DataType dtype => TF_DataType.DtInvalid;
         public string name => _handle == IntPtr.Zero ? null : c_api.StringPiece(c_api.TF_OperationName(_handle));
         public string OpType => _handle == IntPtr.Zero ? null : c_api.StringPiece(c_api.TF_OperationOpType(_handle));
-        public string Device => _handle == IntPtr.Zero ? null : c_api.StringPiece(c_api.TF_OperationDevice(_handle));
-
+        public string Device => _handle == IntPtr.Zero ? null : c_api.StringPiece(c_api.TF_OperationDevice(_handle));
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public NodeDef node_def
         {
             get

src/TensorFlowNET.Core/Operations/array_ops.py.cs

@@ -601,7 +601,16 @@ namespace Tensorflow
         }
 
         public static Tensor gather<T1, T2>(T1 @params, T2 indices, string name = null, int axis = 0)
-            => gen_array_ops.gather_v2(@params, indices, axis, name: name);
+        {
+            if (axis != 0)
+                return gen_array_ops.gather_v2(@params, indices, axis, name: name);
+
+            if (@params is ResourceVariable variable &&
+                indices is Tensor indices_tensor)
+                return variable.sparse_read(indices_tensor, name);
+
+            return gen_array_ops.gather_v2(@params, indices, axis, name: name);
+        }
 
         public static Tensor transpose<T1, T2>(T1 a, T2 perm, string name = "transpose", bool conjugate = false)
         {

src/TensorFlowNET.Core/Operations/gen_resource_variable_ops.cs

@@ -31,7 +31,62 @@ namespace Tensorflow
         {
             var _op = _op_def_lib._apply_op_helper("VarIsInitializedOp", name, new { resource });
 
-            return _op;
+            return _op.output;
+        }
+
+        /// <summary>
+        /// Creates a handle to a Variable resource.
+        /// </summary>
+        /// <param name="dtype"></param>
+        /// <param name="shape"></param>
+        /// <param name="container"></param>
+        /// <param name="shared_name"></param>
+        /// <param name="name"></param>
+        /// <returns></returns>
+        public static Tensor var_handle_op(TF_DataType dtype, TensorShape shape, 
+            string container ="", string shared_name = "", string name = null)
+        {
+            var _op = _op_def_lib._apply_op_helper("VarHandleOp", name, new {
+                dtype,
+                shape,
+                container,
+                shared_name
+            });
+
+            return _op.output;
+        }
+
+        /// <summary>
+        /// Reads the value of a variable.
+        /// </summary>
+        /// <param name="resource"></param>
+        /// <param name="dtype"></param>
+        /// <param name="name"></param>
+        /// <returns></returns>
+        public static Tensor read_variable_op(Tensor resource, TF_DataType dtype, string name = null)
+        {
+            var _op = _op_def_lib._apply_op_helper("ReadVariableOp", name, new
+            {
+                resource,
+                dtype
+            });
+
+            return _op.output;
+        }
+
+        public static Tensor resource_gather(Tensor resource, Tensor indices, TF_DataType dtype, 
+            int batch_dims = 0, bool validate_indices = true, string name = null)
+        {
+            var _op = _op_def_lib._apply_op_helper("ResourceGather", name, new
+            {
+                resource,
+                indices,
+                dtype,
+                batch_dims,
+                validate_indices
+            });
+
+            return _op.output;
         }
     }
 }

src/TensorFlowNET.Core/Operations/resource_variable_ops.cs

@@ -16,6 +16,7 @@
 
 using System;
 using Tensorflow.Framework;
+using static Tensorflow.CppShapeInferenceResult.Types;
 
 namespace Tensorflow
 {
@@ -91,12 +92,80 @@ namespace Tensorflow
                 shape, dtype, shared_name, name, graph_mode, initial_value);
         }
 
+        /// <summary>
+        /// Create a new variable handle, optionally copying in `extra_handle_data`
+        /// </summary>
+        /// <param name="shape"></param>
+        /// <param name="dtype"></param>
+        /// <param name="shared_name"></param>
+        /// <param name="name"></param>
+        /// <param name="graph_mode"></param>
+        /// <param name="extra_handle_data"></param>
+        /// <returns></returns>
         public static Tensor variable_handle_from_shape_and_dtype(TensorShape shape, TF_DataType dtype, 
             string shared_name, string name, bool graph_mode, Tensor extra_handle_data = null)
         {
+            var container = "";// ops.get_default_graph().container;
+            var handle = gen_resource_variable_ops.var_handle_op(shape: shape,
+                dtype: dtype,
+                shared_name: shared_name,
+                name: name,
+                container: container);
+
+            if (extra_handle_data == null)
+                extra_handle_data = handle;
+
+            if (graph_mode)
+            {
+                var full_handle_data = _combine_handle_data(handle, extra_handle_data);
+                _set_handle_shapes_and_types(handle, full_handle_data, graph_mode);
+                return handle;
+            }
+            else
+            {
+                throw new NotImplementedException("");
+            }
+        }
+
+        private static void _set_handle_shapes_and_types(Tensor handle, HandleData full_handle_data, bool graph_mode)
+        {
+
+        }
+
+        /// <summary>
+        /// Concats HandleData from tensors `handle` and `initial_value`.
+        /// </summary>
+        /// <param name="handle"></param>
+        /// <param name="initial_value"></param>
+        /// <returns></returns>
+        private static HandleData _combine_handle_data(Tensor handle, Tensor initial_value)
+        {
+            var variable_handle_data = get_eager_safe_handle_data(initial_value);
+
+            if (initial_value.dtype != dtypes.variant)
+                return variable_handle_data;
+
             throw new NotImplementedException("");
         }
 
+        private static HandleData get_eager_safe_handle_data(Tensor handle)
+        {
+            if(handle == IntPtr.Zero)
+            {
+                var data = new HandleData();
+                data.ShapeAndType.Add(new HandleShapeAndType
+                {
+                    Shape = handle.TensorShape.as_proto(),
+                    Dtype = handle.dtype.as_datatype_enum()
+                });
+                return data;
+            }
+            else
+            {
+                return HandleData.Parser.ParseFrom(handle.BufferToArray());
+            }
+        }
+
         /// <summary>
         /// Represents a future for a read of a variable.
         /// Pretends to be the tensor if anyone looks.

src/TensorFlowNET.Core/Protobuf/CppShapeInference.cs

@@ -0,0 +1,692 @@
+// <auto-generated>
+//     Generated by the protocol buffer compiler.  DO NOT EDIT!
+//     source: tensorflow/python/framework/cpp_shape_inference.proto
+// </auto-generated>
+#pragma warning disable 1591, 0612, 3021
+#region Designer generated code
+
+using pb = global::Google.Protobuf;
+using pbc = global::Google.Protobuf.Collections;
+using pbr = global::Google.Protobuf.Reflection;
+using scg = global::System.Collections.Generic;
+namespace Tensorflow {
+
+  /// <summary>Holder for reflection information generated from tensorflow/python/framework/cpp_shape_inference.proto</summary>
+  public static partial class CppShapeInferenceReflection {
+
+    #region Descriptor
+    /// <summary>File descriptor for tensorflow/python/framework/cpp_shape_inference.proto</summary>
+    public static pbr::FileDescriptor Descriptor {
+      get { return descriptor; }
+    }
+    private static pbr::FileDescriptor descriptor;
+
+    static CppShapeInferenceReflection() {
+      byte[] descriptorData = global::System.Convert.FromBase64String(
+          string.Concat(
+            "CjV0ZW5zb3JmbG93L3B5dGhvbi9mcmFtZXdvcmsvY3BwX3NoYXBlX2luZmVy",
+            "ZW5jZS5wcm90bxIKdGVuc29yZmxvdxoldGVuc29yZmxvdy9jb3JlL2ZyYW1l",
+            "d29yay90eXBlcy5wcm90bxosdGVuc29yZmxvdy9jb3JlL2ZyYW1ld29yay90",
+            "ZW5zb3Jfc2hhcGUucHJvdG8i7QIKF0NwcFNoYXBlSW5mZXJlbmNlUmVzdWx0",
+            "EisKBXNoYXBlGAEgASgLMhwudGVuc29yZmxvdy5UZW5zb3JTaGFwZVByb3Rv",
+            "EkMKC2hhbmRsZV9kYXRhGAQgASgLMi4udGVuc29yZmxvdy5DcHBTaGFwZUlu",
+            "ZmVyZW5jZVJlc3VsdC5IYW5kbGVEYXRhGmYKEkhhbmRsZVNoYXBlQW5kVHlw",
+            "ZRIrCgVzaGFwZRgBIAEoCzIcLnRlbnNvcmZsb3cuVGVuc29yU2hhcGVQcm90",
+            "bxIjCgVkdHlwZRgCIAEoDjIULnRlbnNvcmZsb3cuRGF0YVR5cGUabAoKSGFu",
+            "ZGxlRGF0YRIOCgZpc19zZXQYASABKAgSTgoOc2hhcGVfYW5kX3R5cGUYAiAD",
+            "KAsyNi50ZW5zb3JmbG93LkNwcFNoYXBlSW5mZXJlbmNlUmVzdWx0LkhhbmRs",
+            "ZVNoYXBlQW5kVHlwZUoECAIQA0oECAMQBCJlCh1DcHBTaGFwZUluZmVyZW5j",
+            "ZUlucHV0c05lZWRlZBIcChRpbnB1dF90ZW5zb3JzX25lZWRlZBgBIAMoBRIm",
+            "Ch5pbnB1dF90ZW5zb3JzX2FzX3NoYXBlc19uZWVkZWQYAiADKAVCA/gBAWIG",
+            "cHJvdG8z"));
+      descriptor = pbr::FileDescriptor.FromGeneratedCode(descriptorData,
+          new pbr::FileDescriptor[] { global::Tensorflow.TypesReflection.Descriptor, global::Tensorflow.TensorShapeReflection.Descriptor, },
+          new pbr::GeneratedClrTypeInfo(null, new pbr::GeneratedClrTypeInfo[] {
+            new pbr::GeneratedClrTypeInfo(typeof(global::Tensorflow.CppShapeInferenceResult), global::Tensorflow.CppShapeInferenceResult.Parser, new[]{ "Shape", "HandleData" }, null, null, new pbr::GeneratedClrTypeInfo[] { new pbr::GeneratedClrTypeInfo(typeof(global::Tensorflow.CppShapeInferenceResult.Types.HandleShapeAndType), global::Tensorflow.CppShapeInferenceResult.Types.HandleShapeAndType.Parser, new[]{ "Shape", "Dtype" }, null, null, null),
+            new pbr::GeneratedClrTypeInfo(typeof(global::Tensorflow.CppShapeInferenceResult.Types.HandleData), global::Tensorflow.CppShapeInferenceResult.Types.HandleData.Parser, new[]{ "IsSet", "ShapeAndType" }, null, null, null)}),
+            new pbr::GeneratedClrTypeInfo(typeof(global::Tensorflow.CppShapeInferenceInputsNeeded), global::Tensorflow.CppShapeInferenceInputsNeeded.Parser, new[]{ "InputTensorsNeeded", "InputTensorsAsShapesNeeded" }, null, null, null)
+          }));
+    }
+    #endregion
+
+  }
+  #region Messages
+  public sealed partial class CppShapeInferenceResult : pb::IMessage<CppShapeInferenceResult> {
+    private static readonly pb::MessageParser<CppShapeInferenceResult> _parser = new pb::MessageParser<CppShapeInferenceResult>(() => new CppShapeInferenceResult());
+    private pb::UnknownFieldSet _unknownFields;
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public static pb::MessageParser<CppShapeInferenceResult> Parser { get { return _parser; } }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public static pbr::MessageDescriptor Descriptor {
+      get { return global::Tensorflow.CppShapeInferenceReflection.Descriptor.MessageTypes[0]; }
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    pbr::MessageDescriptor pb::IMessage.Descriptor {
+      get { return Descriptor; }
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public CppShapeInferenceResult() {
+      OnConstruction();
+    }
+
+    partial void OnConstruction();
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public CppShapeInferenceResult(CppShapeInferenceResult other) : this() {
+      shape_ = other.shape_ != null ? other.shape_.Clone() : null;
+      handleData_ = other.handleData_ != null ? other.handleData_.Clone() : null;
+      _unknownFields = pb::UnknownFieldSet.Clone(other._unknownFields);
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public CppShapeInferenceResult Clone() {
+      return new CppShapeInferenceResult(this);
+    }
+
+    /// <summary>Field number for the "shape" field.</summary>
+    public const int ShapeFieldNumber = 1;
+    private global::Tensorflow.TensorShapeProto shape_;
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public global::Tensorflow.TensorShapeProto Shape {
+      get { return shape_; }
+      set {
+        shape_ = value;
+      }
+    }
+
+    /// <summary>Field number for the "handle_data" field.</summary>
+    public const int HandleDataFieldNumber = 4;
+    private global::Tensorflow.CppShapeInferenceResult.Types.HandleData handleData_;
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public global::Tensorflow.CppShapeInferenceResult.Types.HandleData HandleData {
+      get { return handleData_; }
+      set {
+        handleData_ = value;
+      }
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public override bool Equals(object other) {
+      return Equals(other as CppShapeInferenceResult);
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public bool Equals(CppShapeInferenceResult other) {
+      if (ReferenceEquals(other, null)) {
+        return false;
+      }
+      if (ReferenceEquals(other, this)) {
+        return true;
+      }
+      if (!object.Equals(Shape, other.Shape)) return false;
+      if (!object.Equals(HandleData, other.HandleData)) return false;
+      return Equals(_unknownFields, other._unknownFields);
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public override int GetHashCode() {
+      int hash = 1;
+      if (shape_ != null) hash ^= Shape.GetHashCode();
+      if (handleData_ != null) hash ^= HandleData.GetHashCode();
+      if (_unknownFields != null) {
+        hash ^= _unknownFields.GetHashCode();
+      }
+      return hash;
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public override string ToString() {
+      return pb::JsonFormatter.ToDiagnosticString(this);
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public void WriteTo(pb::CodedOutputStream output) {
+      if (shape_ != null) {
+        output.WriteRawTag(10);
+        output.WriteMessage(Shape);
+      }
+      if (handleData_ != null) {
+        output.WriteRawTag(34);
+        output.WriteMessage(HandleData);
+      }
+      if (_unknownFields != null) {
+        _unknownFields.WriteTo(output);
+      }
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public int CalculateSize() {
+      int size = 0;
+      if (shape_ != null) {
+        size += 1 + pb::CodedOutputStream.ComputeMessageSize(Shape);
+      }
+      if (handleData_ != null) {
+        size += 1 + pb::CodedOutputStream.ComputeMessageSize(HandleData);
+      }
+      if (_unknownFields != null) {
+        size += _unknownFields.CalculateSize();
+      }
+      return size;
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public void MergeFrom(CppShapeInferenceResult other) {
+      if (other == null) {
+        return;
+      }
+      if (other.shape_ != null) {
+        if (shape_ == null) {
+          shape_ = new global::Tensorflow.TensorShapeProto();
+        }
+        Shape.MergeFrom(other.Shape);
+      }
+      if (other.handleData_ != null) {
+        if (handleData_ == null) {
+          handleData_ = new global::Tensorflow.CppShapeInferenceResult.Types.HandleData();
+        }
+        HandleData.MergeFrom(other.HandleData);
+      }
+      _unknownFields = pb::UnknownFieldSet.MergeFrom(_unknownFields, other._unknownFields);
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public void MergeFrom(pb::CodedInputStream input) {
+      uint tag;
+      while ((tag = input.ReadTag()) != 0) {
+        switch(tag) {
+          default:
+            _unknownFields = pb::UnknownFieldSet.MergeFieldFrom(_unknownFields, input);
+            break;
+          case 10: {
+            if (shape_ == null) {
+              shape_ = new global::Tensorflow.TensorShapeProto();
+            }
+            input.ReadMessage(shape_);
+            break;
+          }
+          case 34: {
+            if (handleData_ == null) {
+              handleData_ = new global::Tensorflow.CppShapeInferenceResult.Types.HandleData();
+            }
+            input.ReadMessage(handleData_);
+            break;
+          }
+        }
+      }
+    }
+
+    #region Nested types
+    /// <summary>Container for nested types declared in the CppShapeInferenceResult message type.</summary>
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public static partial class Types {
+      public sealed partial class HandleShapeAndType : pb::IMessage<HandleShapeAndType> {
+        private static readonly pb::MessageParser<HandleShapeAndType> _parser = new pb::MessageParser<HandleShapeAndType>(() => new HandleShapeAndType());
+        private pb::UnknownFieldSet _unknownFields;
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public static pb::MessageParser<HandleShapeAndType> Parser { get { return _parser; } }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public static pbr::MessageDescriptor Descriptor {
+          get { return global::Tensorflow.CppShapeInferenceResult.Descriptor.NestedTypes[0]; }
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        pbr::MessageDescriptor pb::IMessage.Descriptor {
+          get { return Descriptor; }
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public HandleShapeAndType() {
+          OnConstruction();
+        }
+
+        partial void OnConstruction();
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public HandleShapeAndType(HandleShapeAndType other) : this() {
+          shape_ = other.shape_ != null ? other.shape_.Clone() : null;
+          dtype_ = other.dtype_;
+          _unknownFields = pb::UnknownFieldSet.Clone(other._unknownFields);
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public HandleShapeAndType Clone() {
+          return new HandleShapeAndType(this);
+        }
+
+        /// <summary>Field number for the "shape" field.</summary>
+        public const int ShapeFieldNumber = 1;
+        private global::Tensorflow.TensorShapeProto shape_;
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public global::Tensorflow.TensorShapeProto Shape {
+          get { return shape_; }
+          set {
+            shape_ = value;
+          }
+        }
+
+        /// <summary>Field number for the "dtype" field.</summary>
+        public const int DtypeFieldNumber = 2;
+        private global::Tensorflow.DataType dtype_ = 0;
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public global::Tensorflow.DataType Dtype {
+          get { return dtype_; }
+          set {
+            dtype_ = value;
+          }
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public override bool Equals(object other) {
+          return Equals(other as HandleShapeAndType);
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public bool Equals(HandleShapeAndType other) {
+          if (ReferenceEquals(other, null)) {
+            return false;
+          }
+          if (ReferenceEquals(other, this)) {
+            return true;
+          }
+          if (!object.Equals(Shape, other.Shape)) return false;
+          if (Dtype != other.Dtype) return false;
+          return Equals(_unknownFields, other._unknownFields);
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public override int GetHashCode() {
+          int hash = 1;
+          if (shape_ != null) hash ^= Shape.GetHashCode();
+          if (Dtype != 0) hash ^= Dtype.GetHashCode();
+          if (_unknownFields != null) {
+            hash ^= _unknownFields.GetHashCode();
+          }
+          return hash;
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public override string ToString() {
+          return pb::JsonFormatter.ToDiagnosticString(this);
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public void WriteTo(pb::CodedOutputStream output) {
+          if (shape_ != null) {
+            output.WriteRawTag(10);
+            output.WriteMessage(Shape);
+          }
+          if (Dtype != 0) {
+            output.WriteRawTag(16);
+            output.WriteEnum((int) Dtype);
+          }
+          if (_unknownFields != null) {
+            _unknownFields.WriteTo(output);
+          }
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public int CalculateSize() {
+          int size = 0;
+          if (shape_ != null) {
+            size += 1 + pb::CodedOutputStream.ComputeMessageSize(Shape);
+          }
+          if (Dtype != 0) {
+            size += 1 + pb::CodedOutputStream.ComputeEnumSize((int) Dtype);
+          }
+          if (_unknownFields != null) {
+            size += _unknownFields.CalculateSize();
+          }
+          return size;
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public void MergeFrom(HandleShapeAndType other) {
+          if (other == null) {
+            return;
+          }
+          if (other.shape_ != null) {
+            if (shape_ == null) {
+              shape_ = new global::Tensorflow.TensorShapeProto();
+            }
+            Shape.MergeFrom(other.Shape);
+          }
+          if (other.Dtype != 0) {
+            Dtype = other.Dtype;
+          }
+          _unknownFields = pb::UnknownFieldSet.MergeFrom(_unknownFields, other._unknownFields);
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public void MergeFrom(pb::CodedInputStream input) {
+          uint tag;
+          while ((tag = input.ReadTag()) != 0) {
+            switch(tag) {
+              default:
+                _unknownFields = pb::UnknownFieldSet.MergeFieldFrom(_unknownFields, input);
+                break;
+              case 10: {
+                if (shape_ == null) {
+                  shape_ = new global::Tensorflow.TensorShapeProto();
+                }
+                input.ReadMessage(shape_);
+                break;
+              }
+              case 16: {
+                dtype_ = (global::Tensorflow.DataType) input.ReadEnum();
+                break;
+              }
+            }
+          }
+        }
+
+      }
+
+      public sealed partial class HandleData : pb::IMessage<HandleData> {
+        private static readonly pb::MessageParser<HandleData> _parser = new pb::MessageParser<HandleData>(() => new HandleData());
+        private pb::UnknownFieldSet _unknownFields;
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public static pb::MessageParser<HandleData> Parser { get { return _parser; } }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public static pbr::MessageDescriptor Descriptor {
+          get { return global::Tensorflow.CppShapeInferenceResult.Descriptor.NestedTypes[1]; }
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        pbr::MessageDescriptor pb::IMessage.Descriptor {
+          get { return Descriptor; }
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public HandleData() {
+          OnConstruction();
+        }
+
+        partial void OnConstruction();
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public HandleData(HandleData other) : this() {
+          isSet_ = other.isSet_;
+          shapeAndType_ = other.shapeAndType_.Clone();
+          _unknownFields = pb::UnknownFieldSet.Clone(other._unknownFields);
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public HandleData Clone() {
+          return new HandleData(this);
+        }
+
+        /// <summary>Field number for the "is_set" field.</summary>
+        public const int IsSetFieldNumber = 1;
+        private bool isSet_;
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public bool IsSet {
+          get { return isSet_; }
+          set {
+            isSet_ = value;
+          }
+        }
+
+        /// <summary>Field number for the "shape_and_type" field.</summary>
+        public const int ShapeAndTypeFieldNumber = 2;
+        private static readonly pb::FieldCodec<global::Tensorflow.CppShapeInferenceResult.Types.HandleShapeAndType> _repeated_shapeAndType_codec
+            = pb::FieldCodec.ForMessage(18, global::Tensorflow.CppShapeInferenceResult.Types.HandleShapeAndType.Parser);
+        private readonly pbc::RepeatedField<global::Tensorflow.CppShapeInferenceResult.Types.HandleShapeAndType> shapeAndType_ = new pbc::RepeatedField<global::Tensorflow.CppShapeInferenceResult.Types.HandleShapeAndType>();
+        /// <summary>
+        /// Only valid if &lt;is_set>.
+        /// </summary>
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public pbc::RepeatedField<global::Tensorflow.CppShapeInferenceResult.Types.HandleShapeAndType> ShapeAndType {
+          get { return shapeAndType_; }
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public override bool Equals(object other) {
+          return Equals(other as HandleData);
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public bool Equals(HandleData other) {
+          if (ReferenceEquals(other, null)) {
+            return false;
+          }
+          if (ReferenceEquals(other, this)) {
+            return true;
+          }
+          if (IsSet != other.IsSet) return false;
+          if(!shapeAndType_.Equals(other.shapeAndType_)) return false;
+          return Equals(_unknownFields, other._unknownFields);
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public override int GetHashCode() {
+          int hash = 1;
+          if (IsSet != false) hash ^= IsSet.GetHashCode();
+          hash ^= shapeAndType_.GetHashCode();
+          if (_unknownFields != null) {
+            hash ^= _unknownFields.GetHashCode();
+          }
+          return hash;
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public override string ToString() {
+          return pb::JsonFormatter.ToDiagnosticString(this);
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public void WriteTo(pb::CodedOutputStream output) {
+          if (IsSet != false) {
+            output.WriteRawTag(8);
+            output.WriteBool(IsSet);
+          }
+          shapeAndType_.WriteTo(output, _repeated_shapeAndType_codec);
+          if (_unknownFields != null) {
+            _unknownFields.WriteTo(output);
+          }
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public int CalculateSize() {
+          int size = 0;
+          if (IsSet != false) {
+            size += 1 + 1;
+          }
+          size += shapeAndType_.CalculateSize(_repeated_shapeAndType_codec);
+          if (_unknownFields != null) {
+            size += _unknownFields.CalculateSize();
+          }
+          return size;
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public void MergeFrom(HandleData other) {
+          if (other == null) {
+            return;
+          }
+          if (other.IsSet != false) {
+            IsSet = other.IsSet;
+          }
+          shapeAndType_.Add(other.shapeAndType_);
+          _unknownFields = pb::UnknownFieldSet.MergeFrom(_unknownFields, other._unknownFields);
+        }
+
+        [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+        public void MergeFrom(pb::CodedInputStream input) {
+          uint tag;
+          while ((tag = input.ReadTag()) != 0) {
+            switch(tag) {
+              default:
+                _unknownFields = pb::UnknownFieldSet.MergeFieldFrom(_unknownFields, input);
+                break;
+              case 8: {
+                IsSet = input.ReadBool();
+                break;
+              }
+              case 18: {
+                shapeAndType_.AddEntriesFrom(input, _repeated_shapeAndType_codec);
+                break;
+              }
+            }
+          }
+        }
+
+      }
+
+    }
+    #endregion
+
+  }
+
+  public sealed partial class CppShapeInferenceInputsNeeded : pb::IMessage<CppShapeInferenceInputsNeeded> {
+    private static readonly pb::MessageParser<CppShapeInferenceInputsNeeded> _parser = new pb::MessageParser<CppShapeInferenceInputsNeeded>(() => new CppShapeInferenceInputsNeeded());
+    private pb::UnknownFieldSet _unknownFields;
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public static pb::MessageParser<CppShapeInferenceInputsNeeded> Parser { get { return _parser; } }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public static pbr::MessageDescriptor Descriptor {
+      get { return global::Tensorflow.CppShapeInferenceReflection.Descriptor.MessageTypes[1]; }
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    pbr::MessageDescriptor pb::IMessage.Descriptor {
+      get { return Descriptor; }
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public CppShapeInferenceInputsNeeded() {
+      OnConstruction();
+    }
+
+    partial void OnConstruction();
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public CppShapeInferenceInputsNeeded(CppShapeInferenceInputsNeeded other) : this() {
+      inputTensorsNeeded_ = other.inputTensorsNeeded_.Clone();
+      inputTensorsAsShapesNeeded_ = other.inputTensorsAsShapesNeeded_.Clone();
+      _unknownFields = pb::UnknownFieldSet.Clone(other._unknownFields);
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public CppShapeInferenceInputsNeeded Clone() {
+      return new CppShapeInferenceInputsNeeded(this);
+    }
+
+    /// <summary>Field number for the "input_tensors_needed" field.</summary>
+    public const int InputTensorsNeededFieldNumber = 1;
+    private static readonly pb::FieldCodec<int> _repeated_inputTensorsNeeded_codec
+        = pb::FieldCodec.ForInt32(10);
+    private readonly pbc::RepeatedField<int> inputTensorsNeeded_ = new pbc::RepeatedField<int>();
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public pbc::RepeatedField<int> InputTensorsNeeded {
+      get { return inputTensorsNeeded_; }
+    }
+
+    /// <summary>Field number for the "input_tensors_as_shapes_needed" field.</summary>
+    public const int InputTensorsAsShapesNeededFieldNumber = 2;
+    private static readonly pb::FieldCodec<int> _repeated_inputTensorsAsShapesNeeded_codec
+        = pb::FieldCodec.ForInt32(18);
+    private readonly pbc::RepeatedField<int> inputTensorsAsShapesNeeded_ = new pbc::RepeatedField<int>();
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public pbc::RepeatedField<int> InputTensorsAsShapesNeeded {
+      get { return inputTensorsAsShapesNeeded_; }
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public override bool Equals(object other) {
+      return Equals(other as CppShapeInferenceInputsNeeded);
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public bool Equals(CppShapeInferenceInputsNeeded other) {
+      if (ReferenceEquals(other, null)) {
+        return false;
+      }
+      if (ReferenceEquals(other, this)) {
+        return true;
+      }
+      if(!inputTensorsNeeded_.Equals(other.inputTensorsNeeded_)) return false;
+      if(!inputTensorsAsShapesNeeded_.Equals(other.inputTensorsAsShapesNeeded_)) return false;
+      return Equals(_unknownFields, other._unknownFields);
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public override int GetHashCode() {
+      int hash = 1;
+      hash ^= inputTensorsNeeded_.GetHashCode();
+      hash ^= inputTensorsAsShapesNeeded_.GetHashCode();
+      if (_unknownFields != null) {
+        hash ^= _unknownFields.GetHashCode();
+      }
+      return hash;
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public override string ToString() {
+      return pb::JsonFormatter.ToDiagnosticString(this);
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public void WriteTo(pb::CodedOutputStream output) {
+      inputTensorsNeeded_.WriteTo(output, _repeated_inputTensorsNeeded_codec);
+      inputTensorsAsShapesNeeded_.WriteTo(output, _repeated_inputTensorsAsShapesNeeded_codec);
+      if (_unknownFields != null) {
+        _unknownFields.WriteTo(output);
+      }
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public int CalculateSize() {
+      int size = 0;
+      size += inputTensorsNeeded_.CalculateSize(_repeated_inputTensorsNeeded_codec);
+      size += inputTensorsAsShapesNeeded_.CalculateSize(_repeated_inputTensorsAsShapesNeeded_codec);
+      if (_unknownFields != null) {
+        size += _unknownFields.CalculateSize();
+      }
+      return size;
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public void MergeFrom(CppShapeInferenceInputsNeeded other) {
+      if (other == null) {
+        return;
+      }
+      inputTensorsNeeded_.Add(other.inputTensorsNeeded_);
+      inputTensorsAsShapesNeeded_.Add(other.inputTensorsAsShapesNeeded_);
+      _unknownFields = pb::UnknownFieldSet.MergeFrom(_unknownFields, other._unknownFields);
+    }
+
+    [global::System.Diagnostics.DebuggerNonUserCodeAttribute]
+    public void MergeFrom(pb::CodedInputStream input) {
+      uint tag;
+      while ((tag = input.ReadTag()) != 0) {
+        switch(tag) {
+          default:
+            _unknownFields = pb::UnknownFieldSet.MergeFieldFrom(_unknownFields, input);
+            break;
+          case 10:
+          case 8: {
+            inputTensorsNeeded_.AddEntriesFrom(input, _repeated_inputTensorsNeeded_codec);
+            break;
+          }
+          case 18:
+          case 16: {
+            inputTensorsAsShapesNeeded_.AddEntriesFrom(input, _repeated_inputTensorsAsShapesNeeded_codec);
+            break;
+          }
+        }
+      }
+    }
+
+  }
+
+  #endregion
+
+}
+
+#endregion Designer generated code

src/TensorFlowNET.Core/TensorFlowNET.Core.csproj

@@ -5,7 +5,7 @@
     <AssemblyName>TensorFlow.NET</AssemblyName>
     <RootNamespace>Tensorflow</RootNamespace>
     <TargetTensorFlow>1.14.0</TargetTensorFlow>
-    <Version>0.11.6</Version>
+    <Version>0.11.8</Version>
     <Authors>Haiping Chen, Meinrad Recheis, Eli Belash</Authors>
     <Company>SciSharp STACK</Company>
     <GeneratePackageOnBuild>true</GeneratePackageOnBuild>
@@ -17,7 +17,7 @@
     <PackageTags>TensorFlow, NumSharp, SciSharp, MachineLearning, TensorFlow.NET, C#</PackageTags>
     <Description>Google's TensorFlow full binding in .NET Standard.
 Docs: https://tensorflownet.readthedocs.io</Description>
-    <AssemblyVersion>0.11.6.0</AssemblyVersion>
+    <AssemblyVersion>0.11.8.0</AssemblyVersion>
     <PackageReleaseNotes>Changes since v0.10.0:
 1. Upgrade NumSharp to v0.20.3.
 2. Add DisposableObject class to manage object lifetime.
@@ -31,9 +31,10 @@ Docs: https://tensorflownet.readthedocs.io</Description>
 10. Support n-dim indexing for tensor.
 11. Add RegisterNoGradients
 12. Add CumsumGrad, BroadcastToGrad.
-13. Return VariableV1 instead of RefVariable.</PackageReleaseNotes>
+13. Return VariableV1 instead of RefVariable.
+14. Add Tensor overload to GradientDescentOptimizer.</PackageReleaseNotes>
     <LangVersion>7.3</LangVersion>
-    <FileVersion>0.11.6.0</FileVersion>
+    <FileVersion>0.11.8.0</FileVersion>
     <PackageLicenseFile>LICENSE</PackageLicenseFile>
     <PackageRequireLicenseAcceptance>true</PackageRequireLicenseAcceptance>
     <SignAssembly>true</SignAssembly>
@@ -42,7 +43,7 @@ Docs: https://tensorflownet.readthedocs.io</Description>
 
   <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|AnyCPU'">
     <AllowUnsafeBlocks>true</AllowUnsafeBlocks>
-    <DefineConstants>TRACE;DEBUG</DefineConstants>
+    <DefineConstants>TRACE;DEBUG;SERIALIZABLE</DefineConstants>
   </PropertyGroup>
 
   <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|AnyCPU'">
@@ -65,6 +66,8 @@ Docs: https://tensorflownet.readthedocs.io</Description>
 
   <ItemGroup>
     <PackageReference Include="Google.Protobuf" Version="3.5.1" />
+    <PackageReference Include="Newtonsoft.Json" Version="12.0.2" />
+    <PackageReference Include="NumSharp" Version="0.20.4" />
   </ItemGroup>
 
   <ItemGroup>
@@ -72,8 +75,4 @@ Docs: https://tensorflownet.readthedocs.io</Description>
     <Folder Include="Keras\Initializers\" />
     <Folder Include="Models\" />
   </ItemGroup>
-
-  <ItemGroup>
-    <ProjectReference Include="..\..\..\NumSharp\src\NumSharp.Core\NumSharp.Core.csproj" />
-  </ItemGroup>
 </Project>

src/TensorFlowNET.Core/Tensors/Tensor.Creation.cs

@@ -25,6 +25,7 @@ using System.Text;
 using NumSharp.Backends;
 using NumSharp.Backends.Unmanaged;
 using static Tensorflow.c_api;
+using Newtonsoft.Json;
 
 namespace Tensorflow
 {
@@ -44,11 +45,17 @@ namespace Tensorflow
         /// <summary>
         ///     True if this Tensor holds data allocated by C#.
         /// </summary>
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public bool IsMemoryOwner => AllocationType >= AllocationType.Marshal;
 
         /// <summary>
         ///     The allocation method used to create this Tensor.
         /// </summary>
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public AllocationType AllocationType { get; protected set; }
 
         /// <summary>

src/TensorFlowNET.Core/Tensors/Tensor.cs

@@ -28,6 +28,7 @@ using NumSharp.Backends;
 using NumSharp.Backends.Unmanaged;
 using NumSharp.Utilities;
 using Tensorflow.Framework;
+using Newtonsoft.Json;
 
 namespace Tensorflow
 {
@@ -43,19 +44,29 @@ namespace Tensorflow
         private readonly int _value_index;
         private TF_Output? _tf_output;
         private readonly TF_DataType _override_dtype;
-
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public int Id => _id;
 
         /// <summary>
         ///     The Graph that contains this tensor.
         /// </summary>
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public Graph graph => op?.graph;
 
         /// <summary>
         ///     The Operation that produces this tensor as an output.
         /// </summary>
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public Operation op => _op;
-
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public Tensor[] outputs => op.outputs;
 
         /// <summary>
@@ -72,24 +83,40 @@ namespace Tensorflow
         ///     The DType of elements in this tensor.
         /// </summary>
         public TF_DataType dtype => _handle == IntPtr.Zero ? _override_dtype : c_api.TF_TensorType(_handle);
-
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public ulong bytesize => _handle == IntPtr.Zero ? 0 : c_api.TF_TensorByteSize(_handle);
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public ulong itemsize => _handle == IntPtr.Zero ? 0 : c_api.TF_DataTypeSize(dtype);
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public ulong size => _handle == IntPtr.Zero ? 0 : bytesize / itemsize;
-        public IntPtr buffer => _handle == IntPtr.Zero ? IntPtr.Zero : c_api.TF_TensorData(_handle);
+        private IntPtr buffer => _handle == IntPtr.Zero ? IntPtr.Zero : c_api.TF_TensorData(_handle);
         public int num_consumers(TF_Output oper_out) => _handle == IntPtr.Zero ? 0 : c_api.TF_OperationOutputNumConsumers(oper_out);
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public int NDims => rank;
 
         /// <summary>
         ///     The name of the device on which this tensor will be produced, or null.
         /// </summary>
         public string Device => op.Device;
-
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public int[] dims => shape;
 
         /// <summary>
         ///     Used for keep other pointer when do implicit operating
         /// </summary>
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public object Tag { get; set; }
 
 
@@ -139,6 +166,9 @@ namespace Tensorflow
             return rank < 0 ? null : shape;
         }
 
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public TensorShape TensorShape => rank < 0 ? new TensorShape() : tensor_util.to_shape(shape);
 
         /// <summary>
@@ -479,9 +509,11 @@ namespace Tensorflow
             } else
                 throw new InvalidOperationException($"Tensor.AllocationHandle is not null ({AllocationHandle}) but AllocationType is not matched to a C# allocation type ({AllocationType}).");
         }
-
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public bool IsDisposed => _disposed;
 
-        public int tensor_int_val { get; set; }
+        // public int tensor_int_val { get; set; }
     }
 }

src/TensorFlowNET.Core/Tensors/TensorShape.cs

@@ -1,4 +1,5 @@
-using NumSharp;
+using Newtonsoft.Json;
+using NumSharp;
 using System;
 using System.Collections.Generic;
 using System.Diagnostics.CodeAnalysis;
@@ -35,6 +36,9 @@ namespace Tensorflow
         /// <summary>
         ///     Returns the size this shape represents.
         /// </summary>
+#if SERIALIZABLE
+        [JsonIgnore]
+#endif
         public int size
         {
             get

src/TensorFlowNET.Core/Tensors/dtypes.cs

@@ -33,6 +33,7 @@ namespace Tensorflow
         public static TF_DataType float32 = TF_DataType.TF_FLOAT; // is that float32?
         public static TF_DataType float16 = TF_DataType.TF_HALF;
         public static TF_DataType float64 = TF_DataType.TF_DOUBLE;
+        public static TF_DataType variant = TF_DataType.TF_VARIANT;
         public static TF_DataType resource = TF_DataType.TF_RESOURCE;
 
         /// <summary>

src/TensorFlowNET.Core/Train/ExponentialMovingAverage.cs

@@ -46,7 +46,7 @@ namespace Tensorflow.Train
                         value,
                         name,
                         colocate_with_primary: true);
-                    ops.add_to_collection(ops.GraphKeys.MOVING_AVERAGE_VARIABLES, var);
+                    ops.add_to_collection(tf.GraphKeys.MOVING_AVERAGE_VARIABLES, var);
                     _averages[var] = avg;
                 }
                 else

src/TensorFlowNET.Core/Train/GradientDescentOptimizer.cs

@@ -39,6 +39,12 @@ namespace Tensorflow.Train
             : base(learning_rate, use_locking, name)
         {
             _lr = learning_rate;
+        }
+
+        public GradientDescentOptimizer(Tensor learning_rate, bool use_locking = false, string name = "GradientDescent")
+            : base(learning_rate, use_locking, name)
+        {
+            _lr_t = learning_rate;
         }
 
         public override void _prepare()

src/TensorFlowNET.Core/Variables/ResourceVariable.cs

@@ -115,6 +115,7 @@ namespace Tensorflow
                         dtype: dtype);
                 });
                 _shape = shape ?? (initial_value as Tensor).TensorShape;
+                _initial_value = initial_value as Tensor;
                 _handle = resource_variable_ops.eager_safe_variable_handle(
                       initial_value: _initial_value,
                       shape: _shape,
@@ -122,7 +123,6 @@ namespace Tensorflow
                       name: name,
                       graph_mode: _in_graph_mode);
                 _unique_id = unique_id;
-                _initial_value = initial_value as Tensor;
                 _handle_name = handle_name + ":0";
                 _dtype = _initial_value.dtype.as_base_dtype();
                 // _constraint = constraint;
@@ -133,6 +133,7 @@ namespace Tensorflow
                     {
                         _is_initialized_op = gen_resource_variable_ops.var_is_initialized_op(_handle);
                     });
+
                     if(initial_value != null)
                     {
                         tf_with(ops.name_scope("Assign"), scope1 =>
@@ -143,10 +144,25 @@ namespace Tensorflow
                                 name: n);
                         });
                     }
+
+                    // Manually assign reads to the handle's device to avoid log
+                    // messages.
+                    tf_with(ops.name_scope("Read"), delegate
+                    {
+                        var value = _read_variable_op();
+                        _graph_element = value;
+                    });
                 }
+
+                ops.add_to_collections(collections, this);
             });
+        }
 
-            throw new NotImplementedException("");
+        private Tensor _read_variable_op()
+        {
+            var result = gen_resource_variable_ops.read_variable_op(_handle, _dtype);
+            // _maybe_set_handle_data(_dtype, _handle, result);
+            return result;
         }
 
         private void _init_from_proto(VariableDef variable_def, string import_scope = null)
@@ -200,6 +216,18 @@ namespace Tensorflow
             _dtype = dtypes.as_tf_dtype((DataType)_handle.op.get_attr("dtype"));
         }
 
+        public Tensor sparse_read(Tensor indices, string name = "Gather")
+        {
+            return tf_with(ops.name_scope(name), scope =>
+            {
+                name = scope;
+                var value = gen_resource_variable_ops.resource_gather(
+                    _handle, indices, dtype: _dtype, name: name);
+
+                return array_ops.identity(value);
+            });
+        }
+
         public override string ToString()
         {
             return $"tf.ResourceVariable '{name}' shape={shape} dtype={dtype}";

src/TensorFlowNET.Core/Variables/variables.py.cs

@@ -135,6 +135,28 @@ namespace Tensorflow
             }
 
             // If at least one input was modified, replace the op.
+            if(modified)
+            {
+                var new_op_type = op_type;
+                if (new_op_type == "RefSwitch")
+                    new_op_type = "Switch";
+                var new_op_name = op.node_def.Name + "_" + name;
+                new_op_name = new_op_name.Replace(":", "_");
+                var _output_types = op._output_types;
+
+                // Convert attr values to AttrValue protos.
+                var attr_protos = new Dictionary<string, AttrValue>();
+                foreach (var attr_def in op.node_def.Attr)
+                    attr_protos[attr_def.Key] = attr_def.Value;
+
+                return op.graph.create_op(
+                    new_op_type, 
+                    new_op_inputs.ToArray(),
+                    _output_types,
+                    name: new_op_name, 
+                    attrs: attr_protos);
+            }
+
             return op;
         }
 

src/TensorFlowNET.Core/ops.GraphKeys.cs

@@ -30,8 +30,10 @@ namespace Tensorflow
         public class GraphKeys
         {
             #region const
-
-            
+            /// <summary>
+            /// Key to collect concatenated sharded variables.
+            /// </summary>
+            public const string CONCATENATED_VARIABLES_ = "concatenated_variables";
             /// <summary>
             /// the subset of `Variable` objects that will be trained by an optimizer.
             /// </summary>
@@ -52,7 +54,12 @@ namespace Tensorflow
             /// </summary>
             public const string LOSSES_ = "losses";
 
-            public const string MOVING_AVERAGE_VARIABLES = "moving_average_variables";
+            public const string LOCAL_VARIABLES_ = "local_variables";
+
+            public const string METRIC_VARIABLES_ = "metric_variables";
+            public const string MODEL_VARIABLES_ = "model_variables";
+
+            public const string MOVING_AVERAGE_VARIABLES_ = "moving_average_variables";
 
             /// <summary>
             /// Key to collect Variable objects that are global (shared across machines).
@@ -64,7 +71,21 @@ namespace Tensorflow
 
             public const string GLOBAL_STEP_ = "global_step";
 
-            public string[] _VARIABLE_COLLECTIONS_ = new string[] { "variables", "trainable_variables", "model_variables" }; 
+            /// <summary>
+            /// List of all collections that keep track of variables.
+            /// </summary>
+            public string[] _VARIABLE_COLLECTIONS_ = new string[] 
+            {
+                GLOBAL_VARIABLES_,
+                LOCAL_VARIABLES_,
+                METRIC_VARIABLES_,
+                MODEL_VARIABLES_,
+                TRAINABLE_VARIABLES_,
+                MOVING_AVERAGE_VARIABLES_,
+                CONCATENATED_VARIABLES_,
+                TRAINABLE_RESOURCE_VARIABLES_
+            }; 
+
             /// <summary>
             /// Key to collect BaseSaverBuilder.SaveableObject instances for checkpointing.
             /// </summary>
@@ -86,7 +107,8 @@ namespace Tensorflow
 
             #endregion
 
-            
+
+            public string CONCATENATED_VARIABLES => CONCATENATED_VARIABLES_;
             /// <summary>
             /// the subset of `Variable` objects that will be trained by an optimizer.
             /// </summary>
@@ -106,13 +128,16 @@ namespace Tensorflow
             /// Key to collect local variables that are local to the machine and are not
             /// saved/restored.
             /// </summary>
-            public string LOCAL_VARIABLES = "local_variables";
+            public string LOCAL_VARIABLES = LOCAL_VARIABLES_;
 
             /// <summary>
             /// Key to collect losses
             /// </summary>
             public string LOSSES => LOSSES_;
 
+            public string METRIC_VARIABLES => METRIC_VARIABLES_;
+            public string MOVING_AVERAGE_VARIABLES = MOVING_AVERAGE_VARIABLES_;
+
             /// <summary>
             /// Key to collect Variable objects that are global (shared across machines).
             /// Default collection for all variables, except local ones.

src/TensorFlowNET.Hub/TensorFlowNET.Hub.csproj

@@ -18,6 +18,6 @@
     <AssemblyName>TensorFlow.Hub</AssemblyName>
   </PropertyGroup>
   <ItemGroup>
-    <ProjectReference Include="..\..\..\NumSharp\src\NumSharp.Core\NumSharp.Core.csproj" />
+    <PackageReference Include="NumSharp" Version="0.20.4" />
   </ItemGroup>
 </Project>

test/TensorFlowNET.UnitTest/MultithreadingTests.cs

@@ -82,8 +82,7 @@ namespace TensorFlowNET.UnitTest
                     var sess_graph = sess.GetPrivate<Graph>("_graph");
                     sess_graph.Should().NotBeNull();
                     default_graph.Should().NotBeNull()
-                        .And.BeEquivalentTo(sess_graph)
-                        .And.BeEquivalentTo(beforehand);
+                        .And.BeEquivalentTo(sess_graph);
 
                     Console.WriteLine($"{tid}-{default_graph.graph_key}");