ubuntu 14.04 - Caffe ImageNet 32X32 images -
ubuntu 14.04 - Caffe ImageNet 32X32 images -
so problem consists of not beingness able train imagenet smaller images (32x32) when resize them 256x256 starts training fine. know issue settings.
i tried set own settings for:
deploy.prototxt: set lastly 2 input_dims 32
solver.prototxt: set solver_mode: cpu (left else alone)
train_val.prototxt: set crop_size: 31 both of settings
all paths right training runs fine images resized 256x256
but setup described above error:
i1114 11:53:38.484948 4566 net.cpp:96] setting fc6 f1114 11:53:38.484967 4566 blob.cpp:72] check failed: data_ *** check failure stack trace: *** @ 0x7f82ed8a0daa (unknown) @ 0x7f82ed8a0ce4 (unknown) @ 0x7f82ed8a06e6 (unknown) @ 0x7f82ed8a3687 (unknown) @ 0x45814e caffe::blob<>::mutable_cpu_data() @ 0x4c12fa caffe::gaussianfiller<>::fill() @ 0x4e5719 caffe::innerproductlayer<>::layersetup() @ 0x4705e1 caffe::net<>::init() @ 0x471eee caffe::net<>::net() @ 0x452560 caffe::solver<>::inittrainnet() @ 0x4537c6 caffe::solver<>::init() @ 0x453926 caffe::solver<>::solver() @ 0x415650 caffe::getsolver<>() @ 0x411b44 train() @ 0x40ef21 main @ 0x7f82ea913ec5 (unknown) @ 0x410697 (unknown) @ (nil) (unknown) train_caffenet_letter.sh: line 4: 4566 aborted (core dumped) ./build/tools/caffe train --solver=models/bvlc_reference_caffenet/solver_letter.prototxt
if please explain other settings should alter in these files, or doing wrong?
ive tried follow tutorial didnt quit it: http://drubiano.github.io/2014/06/18/caffe-custom-data.html
ubuntu-14.04 convolution deep-learning
Comments
Post a Comment