我一直在阅读theano's logistic regression tutorial。我试图理解negative log likelihood是如何计算的。
function drawchart(){
var yaxisdata = [yaxischartdata];
var labeldata = [xaxischartdata];
var Data1 = {
labels : labeldata,
datasets : [{
fillColor : "rgba(151,187,205,0.2)",
strokeColor : "rgba(151,187,205,1)",
pointColor : "rgba(151,187,205,1)",
pointStrokeColor : "#fff",
pointHighlightFill : "#fff",
pointHighlightStroke : "rgba(151,187,205,1)",
data : yaxisdata
}]
};
var ctx = $("#myChart").get(0).getContext("2d");
var myLineChart = new Chart(ctx).Line(Data1);
};
在漂亮的打印打印y = ivector('y')
W = dmatrix('W')
b = dvector('b')
input = dmatrix('inp')
p_y_given_x = T.nnet.softmax(T.dot(input, W) + b)
logs = T.log(self.p_y_given_x)[T.arange(y.shape[0]), y]
上,它返回了
theano.printing.pprint(logs)
有人能用一个小小的例子解释这个'AdvancedSubtensor(log(Softmax(x)), ARange(TensorConstant{0}, Constant{0}[Shape(y)], TensorConstant{1}), y)'
做了什么吗?
在此之后,他们计算了AdvanceSubtensor
任何帮助表示赞赏! :)