xees...@gmail.com
Jan 6, 2016, 3:59:20 AM1/6/16
to Caffe Users
Hi All
I have an RGB image.
I want to learn a certain transformation on the data.
The output is also 3 Channel image.
The network works fine with one channel output.
As soon as, three channel output is given network learns nothing.
I am using euclidean loss, so the label and the the expected output both are three channels.
I have no idea that why should this happen.
The prototxt I am using is given as follows, its simple one which was used for image super resolution
name: "SRCNN"
layer {
name: "data"
type: "HDF5Data"
top: "data"
top: "label"
hdf5_data_param {
source: "train.txt"
batch_size: 128
}
include: { phase: TRAIN }
}
layer {
name: "data"
type: "HDF5Data"
top: "data"
top: "label"
hdf5_data_param {
source: "test.txt"
batch_size: 2
}
include: { phase: TEST }
}
layer {
name: "conv1"
type: "Convolution"
bottom: "data"
top: "conv1"
param {
lr_mult: 1
}
param {
lr_mult: 0.1
}
convolution_param {
num_output: 64
kernel_size: 9
stride: 1
pad: 0
weight_filler {
type: "gaussian"
std: 0.001
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "relu1"
type: "ReLU"
bottom: "conv1"
top: "conv1"
}
layer {
name: "conv2"
type: "Convolution"
bottom: "conv1"
top: "conv2"
param {
lr_mult: 1
}
param {
lr_mult: 0.1
}
convolution_param {
num_output: 32
kernel_size: 1
stride: 1
pad: 0
weight_filler {
type: "gaussian"
std: 0.001
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "relu2"
type: "ReLU"
bottom: "conv2"
top: "conv2"
}
layer {
name: "conv3"
type: "Convolution"
bottom: "conv2"
top: "conv3"
param {
lr_mult: 0.1
}
param {
lr_mult: 0.1
}
convolution_param {
num_output: 3
kernel_size: 5
stride: 1
pad: 0
weight_filler {
type: "gaussian"
std: 0.001
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "loss"
type: "EuclideanLoss"
bottom: "conv3"
bottom: "label"
top: "loss"
}
Any help, thoughts, or pointers would be great.
Thanks
Zeest
Reply all
Reply to author
Forward
0 new messages