innerProductParamer has no normalize.

[maryhh]

hi,the IP has no normalize option .did i miss something? message InnerProductParameter { optional uint32 num_output = 1; // The number of outputs for the layer optional bool bias_term = 2 [default = true]; // whether to have bias terms optional FillerParameter weight_filler = 3; // The filler for the weight optional FillerParameter bias_filler = 4; // The filler for the bias

// The first axis to be lumped into a single inner product computation; // all preceding axes are retained in the output. // May be negative to index from the end (e.g., -1 for the last axis). optional int32 axis = 5 [default = 1]; // Specify whether to transpose the weight matrix or not. // If transpose == true, any operations will be performed on the transpose // of the weight matrix. The weight matrix itself is not going to be transposed // but rather the transfer flag of operations will be toggled accordingly. optional bool transpose = 6 [default = false]; }

[huangyangyu]

@maryhh Thank you for reminding. You could use AMSoftmax's fc layer which has 'normalize' parameter.

[maryhh]

@maryhh Thank you for reminding. You could use AMSoftmax's fc layer which has 'normalize' parameter.

Come back! Again! Thank you, i have solved it. Another question, i dont know how to use the data.txt generated by gen_noise.py to create lmdb including gt_label and nosie_label. https://github.com/huangyangyu/NoiseFace/blob/0b434a2c0eb664ca2af36c3bc619629fb27dcf3f/layer/noise_tolerant_fr_layer.cpp#L150 the prototxt file only has three bottom inputs,how to input the nosie label? https://github.com/huangyangyu/NoiseFace/blob/0b434a2c0eb664ca2af36c3bc619629fb27dcf3f/deploy/resnet20_arcface_train.prototxt#L860 by the way, does the noise_gt mean item['clean']? thank you!

[huangyangyu]

@maryhh The default number of bottoms is 3. The fourth bottom, which indicate whether the sample is clean, is just used to help debugging figure. You can just use three bottoms.