caffe - beginner guide

[Arik Rinberg]

Hi,

I'm trying to implement a DNN within caffe for MNIST and I've come across two questions that I can't find answers for:

1. I'm unsure as to what blob dimensions are output from mnist_train_lmdb. What size are the pictures, how are they given? Is there a way to find out?

2. I'm unsure how to do the function Wx+b. I've found the power function, but i don't want the DNN to learn a different power other than 1. Is this possible?

Bassically the idea is:

input->Wx+b->ReLU->Wx+b->ReLU->output

What would be the best way to learn how to do such a thing?

Thanks,

Arik

[Arik Rinberg]

Also I'd like to know if it's possible to start the layers with randomly generated weights and biases, from a standard deviation.

Thanks!