CODEWITHSUNDEEP
neural-networkcaffe
dontloo
dontloo

Reputation: 10865

Caffe variable batch size

I know if I have the input layer as follows, my network will take in blobs of dimension (1,1,100,100).

layer {
  name: "data"
  type: "Input"
  top: "data"
  input_param {
    shape {
      dim: 1
      dim: 1
      dim: 100
      dim: 100
    }
  }
}

What should I do to make the first dimension (input batch size) variable? so that I can feed in the network batches of different sizes?

Upvotes: 3

Views: 1863

Answers (2)

dontloo
dontloo

Reputation: 10865

in addition to AHA's answer, in c++ it's like

Blob<float>* input_layer = net_->input_blobs()[0];
input_layer->Reshape(batch_size, input_layer->shape(1), input_layer->shape(2), input_layer->shape(3));
net_->Reshape();

Upvotes: 0

Amir
Amir

Reputation: 2415

You can reshape the network before calling the forward() method. So if you want a variable batch_size, you should reshape the everytime. This can be done in any interface you are using (C, python, MATLAB).

In python, it goes like this:

net.blobs['data'].reshape(BATCH_SIZE, CHANNELS, HEIGHT, WIDTH)
net.reshape()
net.forward()

hint: I believe net.reshape() is optional and the network calls this before executing the forward action.

Upvotes: 3

Related Questions