please help!

[wolfworld6]

when i run this code,i got this error: Check failed: bottom[0]->count() == bottom[1]->count() (20 vs. 10) SIGMOID_CROSS_ENTROPY_LOSS layer inputs must have the same count.
how do i solve it?

[A1exy]

@wolfworld6 Same error, do u have solved?

[wolfworld6]

@wolfworld6 Same error, do u have solved?

you need to change the classs to fit the bottom[0]

[A1exy]

@wolfworld6 thanks for your reply, does this repo only support binary classification?

[A1exy]

@wolfworld6 if i hava multi class(more than two class), how can i change the classs to fit the bottom[0].

[wolfworld6]

@wolfworld6 if i hava multi class(more than two class), how can i change the classs to fit the bottom[0].

the ith bottom[0] just has label 0 or 1,means multi-binary classification.you can just use the softmax

[A1exy]

@wolfworld6 ok, thanks!

[1343464520]

@wolfworld6 @A1exy i am doing two classification. and i also have the same error, how to solve it? i try to change the output num of inner product layer from 2 to 1. and it seems to work. but the acc is very low, just 60%... my loss layer is just the same as the author's:
layer {
name: "fc1"
type: "InnerProduct"
bottom: "view_blob1"
top: "fc_blob1"
inner_product_param {
num_output: 2
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "loss_focal"
type: "FocalLoss"
bottom: "fc_blob1"
bottom: "label"
top: "loss_focal"
loss_weight: 10
loss_param{
normalize: true
normalization: FULL
}
}
please help me... thanks!