How to run the code in cpu mode

[ahounkanrin]

Hi,

I have been trying to run DeepLab-Context2 in cpu mode without success. I delated the --gpu flag in the trainer.py file and changed the solver_mode to cpu in the solver files. But whenever a run the code the solver_mode turns back to gpu and I get the following error:

DEEPLABV2-RESNET101
Training1 net voc12/vgg128_noup
Running ./.build_release/tools/caffe.bin train --solver=voc12/config/vgg128_noup/solver_train_aug.prototxt --weights=voc12/model/vgg128_noup/init_DEEPLABV2-RESNET101.caffemodel0
I1102 09:39:37.475340 3237 caffe.cpp:204] Using GPUs 0
F1102 09:39:37.504091 3237 common.cpp:66] Cannot use GPU in CPU-only Caffe: check mode.
*** Check failure stack trace: ***
@ 0x7f945dc6913d google::LogMessage::Fail()
@ 0x7f945dc6afa3 google::LogMessage::SendToLog()
@ 0x7f945dc68ccb google::LogMessage::Flush()
@ 0x7f945dc6b98e google::LogMessageFatal::~LogMessageFatal()
@ 0x7f946129f640 caffe::Caffe::SetDevice()
@ 0x561fd0d8b130 train()
@ 0x561fd0d867eb main
@ 0x7f945a2b52e1 __libc_start_main
@ 0x561fd0d8712a _start
@ (nil) (unknown)
Aborted

Any help will much appreciated. Thanks

[Mdwntr]

Hi，I have a same question,do you solve it?

[ahounkanrin]

Hi @Mdwntr , I haven't solved it.