Working on Examples

Backends/SiaNet.Backend.ArrayFire/Backend/SiaNetBackend.cs

@@ -606,9 +606,9 @@ public Tensor Tile(Tensor x, int n, int axis = 0)
             for (int i = 0; i < dims.Length; i++)
             {
                 if (i == axis)
-                    dims[i] = (uint)n;
-                else
                     dims[i] = 1;
+                else
+                    dims[i] = (uint)n;
             }
 
             return Out(Data.Tile(In(x), dims));

Backends/SiaNet.Backend.CNTK/SiaNetBackend.cs

@@ -124,6 +124,16 @@ public Tensor CreateVariable(float[] data, long[] shape, string name = "")
             return Out(v);
         }
 
+        private Tensor CreateVariable(int[] data, long[] shape, string name = "")
+        {
+            shape = BackendUtil.Row2ColMajor(shape);
+
+            var arr = new CNTK.NDArrayView(BackendUtil.CastShapeInt(shape), Array.ConvertAll(data, x => (float)x), DeviceManager.Current);
+            var v = new CNTK.Variable(BackendUtil.CastShapeInt(shape), VariableKind.Input, CNTK.DataType.Float, arr, false, new AxisVector(), false, name, name);
+
+            return Out(v);
+        }
+
         public Tensor Diag(Tensor x)
         {
             throw new NotImplementedException();
@@ -418,7 +428,6 @@ public Tensor Min(Tensor x, params int[] dims)
                 x = Min(x, CorrDim(x.DimCount, item));
             }
 
-
             return x;
         }
 
@@ -459,7 +468,8 @@ public Tensor Pow(Tensor x, float value)
 
         public void Print(Tensor x, string title = "")
         {
-            BackendUtil.Print(x.Shape, x.ToArray());
+            Console.WriteLine(In(x).GetValue().ToString());
+            //BackendUtil.Print(x.Shape, x.ToArray());
         }
 
         public Tensor RandomBernoulli(long[] shape, float p)
@@ -530,15 +540,12 @@ public Tensor Sinh(Tensor x)
 
         public Tensor SliceCols(Tensor x, long start, long end)
         {
-            end = end + 1;
-            var t = C.Transpose(In(x));
             return Out(C.Slice(In(x), AxisVector.Repeat(new Axis(0), 1), IntVector.Repeat((int)start, 1), IntVector.Repeat((int)end, 1)));
         }
 
         public Tensor SliceRows(Tensor x, long start, long end)
         {
-            end = end + 1;
-            return Out(C.Slice(In(x), AxisVector.Repeat(new Axis(x.DimCount - 1), 1), IntVector.Repeat((int)start, 1), IntVector.Repeat((int)end, 1)));
+            return Out(C.Slice(In(x), new AxisVector(new Axis[] {new Axis(0)}), new IntVector(new int[] { 1, 1, 1, 1, 1, 1 }), new IntVector(new int[] { 3, 3, 3, 3, 3, 3 })));
         }
 
         public Tensor Softmax(Tensor x, int axis = -1)
@@ -613,15 +620,10 @@ public Tensor Tanh(Tensor x)
         public Tensor Tile(Tensor x, int n, int axis = 0)
         {
             axis = CorrDim(x.DimCount, axis);
-
-            uint[] dims = new uint[x.DimCount];
-            for (int i = 0; i < dims.Length; i++)
-            {
-                if (i == axis)
-                    dims[i] = (uint)n;
-                else
-                    dims[i] = 1;
-            }
+            if (axis == 0)
+                axis = 1;
+            else
+                axis = 1;
 
             return Out(C.Splice(VariableVector.Repeat(In(x), n), new Axis(axis)));
         }

Examples/BackendTester/Program.cs

@@ -4,22 +4,20 @@
 using SiaNet.Initializers;
 using System;
 using System.Linq;
-using System.Numerics.Tensors;
 
 namespace BackendTester
 {
     class Program
     {
         static void Main(string[] args)
         {
-            Global.UseEngine(SiaNet.Backend.ArrayFire.SiaNetBackend.Instance, DeviceType.CPU);
+            Global.UseEngine(SiaNet.Backend.CNTKLib.SiaNetBackend.Instance, DeviceType.CPU);
             var K = Global.CurrentBackend;
 
-            var data = new float[] { 1, 2, 3, 4, 5, 6, 7, 8, 9, 1, 2, 3, 4, 5, 6, 7, 8, 9 };
-            var arr = np.array(data);
-            arr = arr.reshape(6, 3);
-
-            var a = K.CreateVariable(new float[] { 1, 2, 3, 4, 5, 6, 7, 8, 9 }, new long[] { 3, 3 });
+            var a = K.CreateVariable(new float[] { 1, 2, 3, 4, 5, 6, 7, 8, 9, 1, 2, 3, 4, 5, 6, 7, 8, 9 }, new long[] { 6, 3 });
+            a.Print();
+
+            var sliced = K.SliceRows(a, 1, 2);
 
             //var d = K.CreateVariable(new float[] { 1 }, new long[] { 1, 1 });
             //c = c + d;

Examples/BasicClassificationWithTitanicDataset/Program.cs

@@ -16,7 +16,7 @@ class Program
         static void Main(string[] args)
         {
             //Setup Engine
-            Global.UseEngine(SiaNet.Backend.MxNetLib.SiaNetBackend.Instance, DeviceType.Default);
+            Global.UseEngine(SiaNet.Backend.TensorSharp.SiaNetBackend.Instance, DeviceType.Default);
 
             var train = LoadTrain(); //Load train data
             var test = LoadTest(); //Load test data
@@ -31,7 +31,7 @@ static void Main(string[] args)
             model.Compile(OptimizerType.Adam, LossType.BinaryCrossEntropy, MetricType.BinaryAccurary);
 
             // Perform training with train and val dataset
-            model.Train(train, epochs: 100, batchSize: 128);
+            model.Train(train, epochs: 100, batchSize: 32);
 
             //var prediction = model.Predict(test);
             //TOps.Round(prediction).Print();

Examples/GettingStarted/Program.cs

@@ -14,7 +14,7 @@ static void Main(string[] args)
         {
             //Setup Engine
 
-            Global.UseEngine(SiaNet.Backend.TensorSharp.SiaNetBackend.Instance, DeviceType.CPU);
+            Global.UseEngine(SiaNet.Backend.MxNetLib.SiaNetBackend.Instance, DeviceType.CPU);
 
             //Prep Data
             var (x, y) = PrepDataset();

SiaNet.Engine/SiaNet.Engine.csproj

@@ -26,7 +26,6 @@
   <ItemGroup>
     <PackageReference Include="System.Drawing.Common" Version="4.5.1" />
     <PackageReference Include="System.Drawing.Primitives" Version="4.3.0" />
-    <PackageReference Include="System.Numerics.Tensors" Version="0.2.0-preview3.19128.7" />
   </ItemGroup>
 
 </Project>

SiaNet.Engine/Tensor.cs

@@ -250,12 +250,12 @@ private static ValueTuple<Tensor, Tensor> BroadcastTensor(Tensor lhs, Tensor rhs
                 {
                     if (lhs.Shape[0] == 1)
                     {
-                        lhs = lhs.RepeatTensor(rhs.Shape[0], 0);
+                        lhs = lhs.RepeatTensor(rhs.Shape[0], 1);
                     }
 
                     if (rhs.Shape[0] == 1)
                     {
-                        rhs = rhs.RepeatTensor(lhs.Shape[0], 0);
+                        rhs = rhs.RepeatTensor(lhs.Shape[0], 1);
                     }
                 }
 

SiaNet/Data/DataFrame.cs

@@ -14,6 +14,8 @@ public class DataFrame
 
         internal NDArray UnderlayingVariable;
 
+        private Tensor UnderlayingTensor;
+
         public DataFrame()
         {
 
@@ -58,7 +60,12 @@ public virtual void Load(params float[] data)
         /// <returns></returns>
         public Tensor GetTensor()
         {
-            return K.CreateVariable(UnderlayingVariable.Data<float>(), Shape);
+            if (UnderlayingTensor == null)
+            {
+                UnderlayingTensor = K.CreateVariable(UnderlayingVariable.Data<float>(), Shape);
+            }
+
+            return UnderlayingTensor;
         }
 
         public Array DataArray
@@ -128,15 +135,18 @@ public Tensor GetBatch(int start, int size, int axis = 0)
 
             if (start + size <= Shape[0])
             {
-                data = UnderlayingVariable[new NDArray(Enumerable.Range(start, size).ToArray())];
+                return K.SliceRows(GetTensor(), start, start + size - 1);
+                //data = UnderlayingVariable[new NDArray(Enumerable.Range(start, size).ToArray())];
             }
             else
             {
-                int count = (int)Shape[0] - start;
-                data = UnderlayingVariable[new NDArray(Enumerable.Range(start, count).ToArray())];
+                //int count = (int)Shape[0] - start;
+                //data = UnderlayingVariable[new NDArray(Enumerable.Range(start, count).ToArray())];
+
+                return K.SliceRows(GetTensor(), start, Shape[0] - 1);
             }
 
-            return K.CreateVariable(data.Data<float>(), BackendUtil.Int2Long(data.shape));
+            //return K.CreateVariable(data.Data<float>(), BackendUtil.Int2Long(data.shape));
         }
 
         /// <summary>
@@ -151,7 +161,7 @@ public void Head(int count = 5, string title = "")
                 count = UnderlayingVariable.shape[0];
             }
 
-            Console.WriteLine(UnderlayingVariable[new NDArray(new int[] { 0, count })].ToString());
+            Console.WriteLine(UnderlayingVariable[new NDArray(Enumerable.Range(0, count - 1).ToArray())].ToString());
         }
     }
 }

SiaNet/SiaNet.csproj

@@ -1,4 +1,4 @@
-<Project Sdk="Microsoft.NET.Sdk">
+<Project Sdk="Microsoft.NET.Sdk">
 
   <PropertyGroup>
     <TargetFramework>netstandard2.0</TargetFramework>
@@ -53,7 +53,6 @@ Code well structured, easy to extend if you would like to extend with new layer,
     <PackageReference Include="Newtonsoft.Json" Version="12.0.1" />
     <PackageReference Include="NumSharp" Version="0.8.3" />
     <PackageReference Include="OpenCvSharp3-AnyCPU" Version="4.0.0.20181129" />
-    <PackageReference Include="System.Numerics.Tensors" Version="0.2.0-preview3.19128.7" />
   </ItemGroup>
 
 </Project>