protected ValueTuple <Tensor, List <Tensor> > CreateVisualEncoder(Tensor visualInput, List <SimpleDenseLayerDef> denseLayers, string scope)
{
//use the same encoder as in UnityML's python codes
Tensor temp;
List <Tensor> returnWeights = new List <Tensor>();
using (Current.K.name_scope(scope))
{
var conv1 = new Conv2D(16, new int[] { 8, 8 }, new int[] { 4, 4 }, use_bias: visualEncoderBias, kernel_initializer: new GlorotUniform(scale: visualEncoderInitialScale), activation: new ELU());
var conv2 = new Conv2D(32, new int[] { 4, 4 }, new int[] { 2, 2 }, use_bias: visualEncoderBias, kernel_initializer: new GlorotUniform(scale: visualEncoderInitialScale), activation: new ELU());
temp = conv1.Call(visualInput)[0];
temp = conv2.Call(temp)[0];
var flatten = new Flatten();
//temp = Current.K.batch_flatten(temp);
temp = flatten.Call(temp)[0];
returnWeights.AddRange(conv1.weights);
returnWeights.AddRange(conv2.weights);
}
var output = BuildSequentialLayers(denseLayers, temp, scope);
var hiddenFlat = output.Item1;
returnWeights.AddRange(output.Item2);
return(ValueTuple.Create(hiddenFlat, returnWeights));
}
public override NDArrayOrSymbol HybridForward(NDArrayOrSymbol x, params NDArrayOrSymbol[] args)
{
var residual = x;
x = bn1.Call(x, args);
if (x.IsNDArray)
{
x = nd.Activation(x.NdX, ActivationType.Relu);
}
else
{
x = sym.Activation(x.SymX, ActivationType.Relu);
}
if (ds != null)
{
residual = ds.Call(x, args);
}
x = conv1.Call(x, args);
x = bn2.Call(x, args);
if (x.IsNDArray)
{
x = nd.Activation(x.NdX, ActivationType.Relu);
}
else
{
x = sym.Activation(x.SymX, ActivationType.Relu);
}
x = conv2.Call(x, args);
x = bn3.Call(x, args);
if (x.IsNDArray)
{
x = nd.Activation(x.NdX, ActivationType.Relu);
}
else
{
x = sym.Activation(x.SymX, ActivationType.Relu);
}
x = conv3.Call(x, args);
if (x.IsNDArray)
{
return(x.NdX + residual.NdX);
}
return(x.SymX + residual.SymX);
}
public void TestConv2D()
{
var inputLayer = UnityTFUtils.Input(shape: new int?[] { 32, 32, 3 });
var conv1 = new Conv2D(16, new int[] { 3, 3 }, padding: PaddingType.Same, activation: new ReLU());
var conv2 = new Conv2D(3, new int[] { 3, 3 }, padding: PaddingType.Same, activation: new ReLU());
var target = UnityTFUtils.Input(shape: new int?[] { 32, 32, 3 });
var pred = conv2.Call(conv1.Call(inputLayer[0])[0])[0];
var lossM = new MeanSquareError();
lossM.Call(target[0], pred);
((UnityTFBackend)K).ExportGraphDef("SavedGraph/convLayer.pb");
}
