keras.Sequential model predict get an exception

[shuizhugeju]

TensorFlow.NET 0.32.0 TensorFlow.Keras 0.3.0 SciSharp.TensorFlow.Redist 2.3.1

            var x = np.array(new float[,] { { 0, 0 }, { 0, 1 }, { 1, 0 }, { 1, 1 } });
            var y = np.array(new float[] { 0, 1, 1, 0 });
            //keras.optimizers.Adam
            var layers = new List<ILayer>() {
                keras.layers.Dense(32,activation:"relu",input_shape:new TensorShape(2)),
                keras.layers.Dense(32,activation:"relu"),

                keras.layers.Dense(64,activation:"relu"),
                keras.layers.Dense(2,activation:"softmax")
            };
            var model = keras.Sequential(layers);
            model.compile(loss: keras.losses.SparseCategoricalCrossentropy(), new Tensorflow.Keras.Optimizers.Adam(), new string[] { "accuracy" });
            model.fit(x, y, batch_size: 2, epochs: 50, verbose: 1);

            //var p = model.predict(x);
            var p = model.predict(x, verbose: 1); //get exception

at Tensorflow.TensorDataset.<>c.<.ctor>b__0_0(Tensor x) at System.Linq.Enumerable.SelectArrayIterator`2.ToArray() at Tensorflow.TensorDataset..ctor(Tensor feature, Tensor label) at Tensorflow.DatasetManager.from_tensor(Tensor features, Tensor labels) at Tensorflow.Keras.Engine.DataAdapters.TensorLikeDataAdapter.slice_inputs(IDatasetV2 indices_dataset, Tensor x, Tensor y) at Tensorflow.Keras.Engine.DataAdapters.TensorLikeDataAdapter..ctor(DataAdapterArgs args) at Tensorflow.Keras.Engine.DataAdapters.DataHandler..ctor(DataHandlerArgs args) at Tensorflow.Keras.Engine.Model.predict(Tensor x, Int32 batch_size, Int32 verbose, Int32 steps, Int32 max_queue_size, Int32 workers, Boolean use_multiprocessing) at TFtest.Program.abc() in C:\Users\binbin\source\repos\TFtest\TFtest\Program.cs:line 60

[Oceania2018]

Please try latest version v0.4.0.