Now AccuracyLayer only computes accuracy, one should use LossLayers to compute loss
authorSergio <sguada@gmail.com>
Sat, 21 Jun 2014 02:55:59 +0000 (19:55 -0700)
committerSergio <sguada@gmail.com>
Sat, 21 Jun 2014 02:56:18 +0000 (19:56 -0700)
commit8d4dacf6d9b5151937927eea34f96ebc6912eb31
treeaeaf05799082e217e506762f3bc8bade0126c9a5
parent03cd25ce658c0270dd8bcf4c4cc0ff206f87aa05
Now AccuracyLayer only computes accuracy, one should use LossLayers to compute loss
Changed all val.prototxt in examples to add a LossLayer to compute loss in Test
examples/cifar10/cifar10_full_test.prototxt
examples/cifar10/cifar10_quick_test.prototxt
examples/feature_extraction/imagenet_val.prototxt
examples/imagenet/alexnet_val.prototxt
examples/imagenet/imagenet_val.prototxt
examples/mnist/lenet_consolidated_solver.prototxt
examples/mnist/lenet_test.prototxt
examples/mnist/mnist_autoencoder_test.prototxt
examples/pascal-finetuning/pascal_finetune_val.prototxt
include/caffe/loss_layers.hpp
src/caffe/layers/accuracy_layer.cpp