"comment":"\nHardSigmoid Activation Operator.\n\nSegment-wise linear approximation of sigmoid(https://arxiv.org/abs/1603.00391), \nwhich is much faster than sigmoid.\n\n$y = \\max(0, \\min(1, slope * x + shift))$\n\nThe slope should be positive. The offset can be either positive or negative.\nThe default slope and shift are set according to the above reference.\nIt is recommended to use the defaults for this activation.\n\n",
"comment":"\nHardSigmoid Activation Operator.\n\nSegment-wise linear approximation of sigmoid(https://arxiv.org/abs/1603.00391), \nwhich is much faster than sigmoid.\n\n$out = \\max(0, \\min(1, slope * x + shift))$\n\nThe slope should be positive. The offset can be either positive or negative.\nThe default slope and shift are set according to the above reference.\nIt is recommended to use the defaults for this activation.\n\n",
"comment":"\nELU Activation Operator.\n\nApplies the following element-wise computation on the input according to\nhttps://arxiv.org/abs/1511.07289.\n\n$y = \\max(0, x) + \\min(0, \\alpha * (e^x - 1))$\n\n",
"comment":"\nELU Activation Operator.\n\nApplies the following element-wise computation on the input according to\nhttps://arxiv.org/abs/1511.07289.\n\n$out = \\max(0, x) + \\min(0, \\alpha * (e^x - 1))$\n\n",
"comment":"\nSoftmax Operator.\n\nThe input of the softmax operator is a 2-D tensor with shape N x K (N is the\nbatch_size, K is the dimension of input feature). The output tensor has the\nsame shape as the input tensor.\n\nFor each row of the input tensor, the softmax operator squashes the\nK-dimensional vector of arbitrary real values to a K-dimensional vector of real\nvalues in the range [0, 1] that add up to 1.\nIt computes the exponential of the given dimension and the sum of exponential\nvalues of all the other dimensions in the K-dimensional vector input.\nThen the ratio of the exponential of the given dimension and the sum of\nexponential values of all the other dimensions is the output of the softmax\noperator.\n\nFor each row $i$ and each column $j$ in Input(X), we have:\n $$Y[i, j] = \\frac{\\exp(X[i, j])}{\\sum_j(exp(X[i, j])}$$\n\n",
"comment":"\nSoftmax Operator.\n\nThe input of the softmax operator is a 2-D tensor with shape N x K (N is the\nbatch_size, K is the dimension of input feature). The output tensor has the\nsame shape as the input tensor.\n\nFor each row of the input tensor, the softmax operator squashes the\nK-dimensional vector of arbitrary real values to a K-dimensional vector of real\nvalues in the range [0, 1] that add up to 1.\nIt computes the exponential of the given dimension and the sum of exponential\nvalues of all the other dimensions in the K-dimensional vector input.\nThen the ratio of the exponential of the given dimension and the sum of\nexponential values of all the other dimensions is the output of the softmax\noperator.\n\nFor each row $i$ and each column $j$ in Input(X), we have:\n $$Out[i, j] = \\frac{\\exp(X[i, j])}{\\sum_j(exp(X[i, j])}$$\n\n",
"inputs":[
"inputs":[
{
{
"name":"X",
"name":"X",
...
@@ -1022,7 +1022,7 @@
...
@@ -1022,7 +1022,7 @@
}],
}],
"outputs":[
"outputs":[
{
{
"name":"Y",
"name":"Out",
"comment":"The normalized values with the same shape as X.",
"comment":"The normalized values with the same shape as X.",
