FazBrowse GitHub Viewer | Trending |
URL:
| Home
Tools: [Download Repo ZIP]   [Original HTTPS Page]

np.average #894 · MSavameri/TensorFlow.NET@a22e92d · GitHub

  1. MSavameri
  2. TensorFlow.NET

Repository navigation

Commit a22e92d

Browse files
Oceania2018
committed
np.average SciSharp#894 
1 parent 4b4f7f8 commit a22e92d

6 files changed

Lines changed: 153 additions & 1 deletion

File tree

‎src/TensorFlowNET.Core/NumPy/Implementation/NumPyImpl.Statistics.cs‎

Lines changed: 31 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,31 @@
1+
using System;
2+
using System.Collections.Generic;
3+
4+
namespace Tensorflow.NumPy
5+
{
6+
public partial class NumPyImpl
7+
{
8+
public NDArray average(NDArray a, int axis = -1, NDArray? weights = null, bool returned = false)
9+
{
10+
var dtype = NumPyUtils.GetResultType(a.dtype, np.float64);
11+
if(weights is null)
12+
{
13+
var tensorA = math_ops.cast(a, dtype);
14+
var nd = math_ops.reduce_mean(tensorA, axis);
15+
return new NDArray(nd);
16+
}
17+
else
18+
{
19+
var tensorW = math_ops.cast(weights, dtype);
20+
if(a.rank != weights.rank)
21+
{
22+
var weights_sum = math_ops.reduce_sum(tensorW);
23+
var axes = ops.convert_to_tensor(new[,] { { axis }, { 0 } });
24+
var avg = math_ops.tensordot(a, weights, axes) / weights_sum;
25+
}
26+
27+
throw new NotImplementedException("");
28+
}
29+
}
30+
}
31+
}

‎src/TensorFlowNET.Core/NumPy/NumPy.Statistics.cs‎

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -14,5 +14,9 @@ public partial class np
1414

1515
[AutoNumPy]
1616
public static NDArray amax(NDArray x, int axis = 0) => new NDArray(tf.math.argmax(x, axis));
17+
18+
[AutoNumPy]
19+
public static NDArray average(NDArray a, int axis = -1, NDArray? weights = null, bool returned = false)
20+
=> tf.numpy.average(a, axis: axis, weights: weights, returned: returned);
1721
}
1822
}

‎src/TensorFlowNET.Core/NumPy/NumPyUtils.cs‎

Lines changed: 19 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,19 @@
1+
using System;
2+
using System.Text;
3+
4+
namespace Tensorflow.NumPy
5+
{
6+
internal class NumPyUtils
7+
{
8+
public static TF_DataType GetResultType(params TF_DataType[] dtypes)
9+
{
10+
var resultDType = dtypes[0];
11+
for(int i = 1; i < dtypes.Length; i++)
12+
{
13+
if (dtypes[i].get_datatype_size() > resultDType.get_datatype_size())
14+
resultDType = dtypes[i];
15+
}
16+
return resultDType;
17+
}
18+
}
19+
}

‎src/TensorFlowNET.Core/Operations/math_ops.cs‎

Lines changed: 66 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -929,6 +929,72 @@ Tensor _tensordot_reshape(Tensor a, int[] axes, bool flipped = false)
929929
throw new NotImplementedException("tensordot");
930930
}
931931

932+
public static Tensor tensordot(Tensor x, Tensor y, Tensor axes, string name = null)
933+
{
934+
Tensor _tensordot_reshape(Tensor a, int[] axes, bool flipped = false)
935+
{
936+
if (a.shape.IsFullyDefined && isinstance(axes, (typeof(List<object>), typeof(Tuple))))
937+
{
938+
var shape_a = a.shape.dims;
939+
940+
// axes
941+
int iter = 0;
942+
foreach (int i in axes)
943+
{
944+
if (i >= 0)
945+
axes[0 + iter] = i;
946+
else
947+
axes[0 + iter] = i + len(shape_a);
948+
iter++;
949+
}
950+
951+
// free
952+
int[] free = { };
953+
iter = 0;
954+
foreach (int i in Enumerable.Range(0, len(axes)))
955+
if (!Array.Exists(axes, i => i == i))
956+
free[free.Length] = i;
957+
958+
// free_dims
959+
int[] free_dims = { };
960+
foreach (int i in free)
961+
free_dims[free_dims.Length] = (int)shape_a[i];
962+
963+
int prod_free = (int)np.prod(free_dims);
964+
965+
// prod_axes
966+
int[] prod_axes_pre = { };
967+
foreach (int i in axes)
968+
prod_axes_pre[prod_axes_pre.Length] = (int)shape_a[i];
969+
int prod_axes = (int)np.prod(prod_axes_pre);
970+
971+
// perm
972+
Tensor perm;
973+
if (flipped)
974+
perm = ops.convert_to_tensor(list(free)) + ops.convert_to_tensor(free);
975+
else
976+
perm = ops.convert_to_tensor(list(free)) + ops.convert_to_tensor(free)
977+
+ ops.convert_to_tensor(list(axes));
978+
979+
// new_shape
980+
Shape new_shape;
981+
if (flipped)
982+
new_shape = new Shape(new int[] { prod_axes, prod_free });
983+
else
984+
new_shape = new Shape(new int[] { prod_free, prod_axes });
985+
}
986+
987+
throw new NotImplementedException("_tensordot_reshape");
988+
}
989+
990+
return tf_with(ops.name_scope(name, "Tensordot", new { x, y, axes }), scope =>
991+
{
992+
name = scope;
993+
var (a_axes, b_axes) = (axes[0], axes[1]);
994+
return x;
995+
});
996+
}
997+
932998
public static Tensor truediv(Tensor x, Tensor y, string name = null)
933999
=> _truediv_python3(x, y, name);
9341000

‎src/TensorFlowNET.Core/Tensors/Tensor.cs‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -78,7 +78,7 @@ public partial class Tensor : DisposableObject,
7878
/// <summary>
7979
/// The name of the device on which this tensor will be produced, or null.
8080
/// </summary>
81-
public virtual string Device => op.Device;
81+
public virtual string Device => op?.Device;
8282
public long[] dims => shape.dims;
8383

8484
/// <summary>

‎test/TensorFlowNET.UnitTest/NumPy/Statistics.Test.cs‎

Lines changed: 32 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,32 @@
1+
using Microsoft.VisualStudio.TestTools.UnitTesting;
2+
using System;
3+
using System.Collections.Generic;
4+
using System.Linq;
5+
using System.Text;
6+
using Tensorflow;
7+
using Tensorflow.NumPy;
8+
9+
namespace TensorFlowNET.UnitTest.NumPy
10+
{
11+
/// <summary>
12+
/// https://numpy.org/doc/stable/reference/routines.statistics.html
13+
/// </summary>
14+
[TestClass]
15+
public class StatisticsTest : EagerModeTestBase
16+
{
17+
[TestMethod]
18+
public void average()
19+
{
20+
var data = np.arange(1, 5);
21+
var avg = np.average(data);
22+
Assert.AreEqual(avg, 2.5);
23+
24+
data = np.arange(6).reshape((3, 2));
25+
avg = np.average(data, axis: 1);
26+
assertAllEqual(avg.ToArray<double>(), new[] { 0.5, 2.5, 4.5 });
27+
28+
// avg = np.average(data, axis: 1, weights: new[] { 1.0 / 4, 3.0 / 4 });
29+
// assertAllEqual(avg.ToArray<double>(), new[] { 0.75, 2.75, 4.75 });
30+
}
31+
}
32+
}

0 commit comments

Comments
 (0)
Back | FazBrowse Home | New Git URL