IEnumerable<NDArray>

Author: Oceania2018

src/TensorFlowNET.Core/Binding.Util.cs

@@ -299,8 +299,8 @@ public static float time()
             where T1 : unmanaged
             where T2 : unmanaged
         {
-            var a = t1.AsIterator<T1>();
-            var b = t2.AsIterator<T2>();
+            //var a = t1.AsIterator<T1>();
+            //var b = t2.AsIterator<T2>();
             //while (a.HasNext() && b.HasNext())
             //yield return (a.MoveNext(), b.MoveNext());
             throw new NotImplementedException("");

src/TensorFlowNET.Core/NumPy/NumPy.Logical.cs

@@ -9,6 +9,8 @@ namespace Tensorflow.NumPy
 {
     public partial class np
     {
+        [AutoNumPy]
+        public static NDArray any(NDArray a, Axis axis = null) => throw new NotImplementedException("");
         [AutoNumPy]
         public static NDArray logical_or(NDArray x1, NDArray x2) => new NDArray(tf.logical_or(x1, x2));
 

src/TensorFlowNET.Core/NumPy/Numpy.Manipulation.cs

@@ -17,6 +17,9 @@ public partial class np
         [AutoNumPy]
         public static NDArray squeeze(NDArray x1, Axis? axis = null) => new NDArray(array_ops.squeeze(x1, axis));
 
+        [AutoNumPy]
+        public static NDArray stack(NDArray arrays, Axis axis = null) => new NDArray(array_ops.stack(arrays, axis ?? 0));
+
         [AutoNumPy]
         public static NDArray dstack(params NDArray[] tup) => throw new NotImplementedException("");
     }

src/TensorFlowNET.Core/NumPy/Numpy.Math.cs

@@ -12,6 +12,9 @@ public partial class np
         [AutoNumPy]
         public static NDArray exp(NDArray x) => new NDArray(tf.exp(x));
 
+        [AutoNumPy]
+        public static NDArray floor(NDArray x) => new NDArray(tf.floor(x));
+
         [AutoNumPy]
         public static NDArray log(NDArray x) => new NDArray(tf.log(x));
 

src/TensorFlowNET.Core/Numpy/NDArray.cs

@@ -15,29 +15,21 @@ limitations under the License.
 ******************************************************************************/
 
 using System;
+using System.Collections;
 using System.Collections.Generic;
 using System.Linq;
 using System.Text;
-using Tensorflow.Eager;
 using static Tensorflow.Binding;
 
 namespace Tensorflow.NumPy
 {
-    public partial class NDArray : Tensor
+    public partial class NDArray : Tensor, IEnumerable<NDArray>
     {
         public IntPtr data => TensorDataPointer;
 
-        public NDArray[] GetNDArrays()
-            => throw new NotImplementedException("");
-
         public ValueType GetValue(params int[] indices)
             => throw new NotImplementedException("");
 
-        public NDIterator<T> AsIterator<T>(bool autoreset = false) where T : unmanaged
-            => throw new NotImplementedException("");
-
-        public bool HasNext() => throw new NotImplementedException("");
-        public T MoveNext<T>() => throw new NotImplementedException("");
         [AutoNumPy]
         public NDArray reshape(Shape newshape) => new NDArray(tf.reshape(this, newshape));
         public NDArray astype(TF_DataType dtype) => new NDArray(math_ops.cast(this, dtype));
@@ -46,5 +38,14 @@ public NDIterator<T> AsIterator<T>(bool autoreset = false) where T : unmanaged
         public Array ToMuliDimArray<T>() => throw new NotImplementedException("");
         public byte[] ToByteArray() => BufferToArray();
         public override string ToString() => NDArrayRender.ToString(this);
+
+        public IEnumerator<NDArray> GetEnumerator()
+        {
+            for (int i = 0; i < dims[0]; i++)
+                yield return this[i];
+        }
+
+        IEnumerator IEnumerable.GetEnumerator()
+            => GetEnumerator();
     }
 }

src/TensorFlowNET.Core/Numpy/NDIterator.Generic.cs

@@ -96,6 +96,12 @@ public static implicit operator Shape((int, int, int, int) dims)
         public static implicit operator Shape((long, long, long, long) dims)
             => new Shape(dims.Item1, dims.Item2, dims.Item3, dims.Item4);
 
+        public static implicit operator Shape((int, int, int, int, int) dims)
+            => new Shape(dims.Item1, dims.Item2, dims.Item3, dims.Item4, dims.Item5);
+
+        public static implicit operator Shape((long, long, long, long, long) dims)
+            => new Shape(dims.Item1, dims.Item2, dims.Item3, dims.Item4, dims.Item5);
+
         public static implicit operator int[](Shape shape)
             => shape.dims.Select(x => (int)x).ToArray();
 

src/TensorFlowNET.Core/Numpy/NDIterator.cs

@@ -1008,7 +1008,7 @@ public static Tensor pad(Tensor tensor, Tensor paddings, string mode = "CONSTANT
                     !(paddings_constant is null))
                 {
                     var new_shape = new List<int>();
-                    foreach ((NDArray padding, int dim) in zip(paddings_constant.GetNDArrays(), np.array(input_shape.dims).GetNDArrays()))
+                    foreach ((NDArray padding, int dim) in zip(paddings_constant, input_shape.as_int_list()))
                     {
                         if (padding is null || dim == -1 || padding.ToArray<int>().Contains(-1))
                             new_shape.Add(-1);

src/TensorFlowNET.Core/Numpy/Shape.cs

@@ -82,7 +82,8 @@ public void Autocast_Case1()
             sess.run(tf.global_variables_initializer());
             var ret = sess.run(op, feed_dict: (input, np.array(1, 2, 3, 4, 5, 6)));
 
-            ret.Should().BeShaped(2, 3).And.BeOfValues(1, 2, 3, 4, 5, 6);
+            Assert.AreEqual(ret.shape, (2, 3));
+            Assert.AreEqual(ret, new[] { 1, 2, 3, 4, 5, 6 });
             print(ret.dtype);
             print(ret);
         }
@@ -110,7 +111,8 @@ public void Autocast_Case3()
             sess.run(tf.global_variables_initializer());
             var ret = sess.run(op, feed_dict: (input, np.array(1, 2, 3, 4, 5, 6).astype(np.float32) + 0.1f));
 
-            ret.Should().BeShaped(2, 3).And.BeOfValues(1, 2, 3, 4, 5, 6);
+            Assert.AreEqual(ret.shape, (2, 3));
+            Assert.AreEqual(ret, new[] { 1, 2, 3, 4, 5, 6 });
             print(ret.dtype);
             print(ret);
         }
@@ -124,7 +126,8 @@ public void Autocast_Case4()
             sess.run(tf.global_variables_initializer());
             var ret = sess.run(op, feed_dict: (input, np.array(1, 2, 3, 4, 5, 6).astype(np.float32) + 0.1f));
 
-            ret.Should().BeShaped(2, 3).And.BeOfValues(1, 2, 3, 4, 5, 6);
+            Assert.AreEqual(ret.shape, (2, 3));
+            Assert.AreEqual(ret, new[] { 1, 2, 3, 4, 5, 6 });
             print(ret.dtype);
             print(ret);
         }