issues
search
281 results
for mini-batch-gradient-descent
Best match
Best match
Most commented
Newest
Recently updated
Least commented
Oldest
Least recently updated
Theano/Theano
#936
SoftmaxGrad should be optimized away
SoftmaxGrad is a generic implementation of the backprop of a gradient through Softmax. Usually, the output of softmax is fed into a cross-entropy cost, and the gradient of that cost wrt the input of s…
lamblin
updated
11 years ago
2
上一页
1
...
23
24
25
26
27
28
29
...
29
下一页