Error "MULTI_STAGE_MEANFIELD" and Failed to parse NetParameter file

[matthiasmace]

Dear Martin
In am willing to make the training run on MacOS. I am able to use CRFasRNN but when I want I run solve.py, I get the following errors :
WARNING: Logging before InitGoogleLogging() is written to STDERR
I1107 23:19:23.737943 3726271424 solver.cpp:54] Initializing solver from parameters:
test_iter: 261
test_interval: 1333
base_lr: 1e-13
display: 50
max_iter: 100000
lr_policy: "fixed"
momentum: 0.99
weight_decay: 0.0005
snapshot: 1000
snapshot_prefix: "models/train"
solver_mode: GPU
net: "TVG_CRFRNN_COCO_VOC_TRAIN_3_CLASSES.prototxt"
test_initialization: false
I1107 23:19:23.738425 3726271424 solver.cpp:96] Creating training net from net file: TVG_CRFRNN_COCO_VOC_TRAIN_3_CLASSES.prototxt
[libprotobuf ERROR google/protobuf/text_format.cc:287] Error parsing text-format caffe.NetParameter: 224:3: Unknown enumeration value of "MULTI_STAGE_MEANFIELD" for field "type".
F1107 23:19:23.739708 3726271424 upgrade_proto.cpp:932] Check failed: ReadProtoFromTextFile(param_file, param) Failed to parse NetParameter file: TVG_CRFRNN_COCO_VOC_TRAIN_3_CLASSES.prototxt
*** Check failure stack trace: ***

Any idea please ?

[Sam813]

hi,
first, check if you install correct caffe version which has the MULTI_STAGE... layer.
second, consider checking the prototext. It is written in the older caffe version, for instance the layers must be replaced with layer, also CONVOLUTION with "Convolution".
a better way is to cross check it with network prototext (TVG_CRFRNN_new_deploy.prototxt)