(var trainX, var trainY, var testX, var testY) = ImportData.IrisDataset<Type>(ratio: 0.8);
Console.WriteLine($"Train on {trainX.Shape[0]}; Test on {testX.Shape[0]}");

var net = new Network<Type>(new SGD<Type>(lr: 0.05), new SquareLoss<Type>(), new ArgmaxAccuracy<Type>());
net.AddLayer(new DenseLayer<Type>(inputNodes: 4, outputNodes: 5, new TanhActivation<Type>()));
net.AddLayer(new DenseLayer<Type>(outputNodes: 3, new SigmoidActivation<Type>()));

net.Summary();

var sw = Stopwatch.StartNew();
net.Fit(trainX, trainY, epochs: 50, batchSize: 10, displayEpochs: 5);
Console.WriteLine($"Time:{sw.ElapsedMilliseconds} ms");

net.Test(testX, testY);

Hello World! Iris MLP. Network<Double>
Train on 120; Test on 30
Summary
Input  Shape:4
Layer: DenseLayer-Tanh      Parameters:    25 Nodes[In: 4 -> Out:5]
Layer: DenseLayer-Sigmoid   Parameters:    18 Nodes[In: 5 -> Out:3]
Output Shape:3
Total Parameters:43

Start Training...
Epochs     0/50 Loss:0.122382 Acc:0.4083
Epochs     5/50 Loss:0.068604 Acc:0.7250
Epochs    10/50 Loss:0.051477 Acc:0.8667
Epochs    15/50 Loss:0.042436 Acc:0.8833
Epochs    20/50 Loss:0.038008 Acc:0.9417
Epochs    25/50 Loss:0.032516 Acc:0.9417
Epochs    30/50 Loss:0.029528 Acc:0.9333
Epochs    35/50 Loss:0.025101 Acc:0.9333
Epochs    40/50 Loss:0.022236 Acc:0.9500
Epochs    45/50 Loss:0.020407 Acc:0.9667
Epochs    50/50 Loss:0.019197 Acc:0.9583
End Training.
Time:195 ms
TestResult Loss:0.009323 Acc:1.0000

(var trainX, var trainY, var testX, var testY) = ImportData.DigitsDataset<Type>(ratio: 0.9);
Console.WriteLine($"Train on {trainX.Shape[0]}; Test on {testX.Shape[0]}");

var net = new Network<Type>(new SGD<Type>(lr: 0.05), new SquareLoss<Type>(), new ArgmaxAccuracy<Type>());
net.AddLayer(new DenseLayer<Type>(inputNodes: 64, outputNodes: 32, new TanhActivation<Type>()));
net.AddLayer(new DenseLayer<Type>(outputNodes: 10, new SigmoidActivation<Type>()));

net.Summary();

var sw = Stopwatch.StartNew();
net.Fit(trainX, trainY, epochs: 50, batchSize: 100, displayEpochs: 5);
Console.WriteLine($"Time:{sw.ElapsedMilliseconds} ms");

net.Test(testX, testY);

Hello World! Digits MLP. Network<Double>
Train on 1617; Test on 180
Summary
Input  Shape:64
Layer: DenseLayer-Tanh      Parameters:  2080 Nodes[In:64 -> Out:32]
Layer: DenseLayer-Sigmoid   Parameters:   330 Nodes[In:32 -> Out:10]
Output Shape:10
Total Parameters:2410

Start Training...
Epochs     0/50 Loss:0.053108 Acc:0.1119
Epochs     5/50 Loss:0.040126 Acc:0.2788
Epochs    10/50 Loss:0.015562 Acc:0.7638
Epochs    15/50 Loss:0.004862 Acc:0.9600
Epochs    20/50 Loss:0.002789 Acc:0.9781
Epochs    25/50 Loss:0.002397 Acc:0.9844
Epochs    30/50 Loss:0.001810 Acc:0.9875
Epochs    35/50 Loss:0.001425 Acc:0.9919
Epochs    40/50 Loss:0.001136 Acc:0.9950
Epochs    45/50 Loss:0.001046 Acc:0.9956
Epochs    50/50 Loss:0.000881 Acc:0.9981
End Training.
Time:21406 ms
TestResult Loss:0.004223 Acc:0.9500