about a training error

[zkh123456789]

when I train MSCNN model ,for example, I input "
cd $MSCNN_ROOT/examples/kitti_car/mscnn-7s-576-2x/
sh train_mscnn.sh
"
then it wil show me that:
I1223 16:55:35.672600 16628 caffe.cpp:204] Using GPUs 0
I1223 16:55:35.691526 16628 caffe.cpp:209] GPU 0: GeForce GTX 1080
I1223 16:55:35.883790 16628 solver.cpp:45] Initializing solver from parameters:
test_iter: 1000
test_interval: 1000
base_lr: 5e-05
display: 50
max_iter: 10000
lr_policy: "step"
gamma: 0.1
momentum: 0.9
weight_decay: 0.0005
stepsize: 10000
snapshot: 10000
snapshot_prefix: "mscnn_kitti_train_1st"
solver_mode: GPU
device_id: 0
random_seed: 1704
net: "trainval_1st.prototxt"
train_state {
level: 0
stage: ""
}
weights: "../../../models/VGG/VGG_ILSVRC_16_layers.caffemodel"
I1223 16:55:35.893198 16628 solver.cpp:102] Creating training net from net file: trainval_1st.prototxt
[libprotobuf ERROR google/protobuf/text_format.cc:274] Error parsing text-format caffe.NetParameter: 22:23: Message type "caffe.LayerParameter" has no field named "image_gt_data_param".
F1223 16:55:35.893450 16628 upgrade_proto.cpp:90] Check failed: ReadProtoFromTextFile(param_file, param) Failed to parse NetParameter file: trainval_1st.prototxt
*** Check failure stack trace: ***
@ 0x7f1517c695cd google::LogMessage::Fail()
@ 0x7f1517c6b433 google::LogMessage::SendToLog()
@ 0x7f1517c6915b google::LogMessage::Flush()
@ 0x7f1517c6be1e google::LogMessageFatal::~LogMessageFatal()
@ 0x7f1518427bf1 caffe::ReadNetParamsFromTextFileOrDie()
@ 0x7f15182a50c5 caffe::Solver<>::InitTrainNet()
@ 0x7f15182a6425 caffe::Solver<>::Init()
@ 0x7f15182a673f caffe::Solver<>::Solver()
@ 0x7f15184432b1 caffe::Creator_SGDSolver<>()
@ 0x40a778 train()
@ 0x407568 main
@ 0x7f15163ff830 __libc_start_main
@ 0x407e39 _start
@ (nil) (unknown)
Aborted (core dumped)
I1223 16:55:36.015866 16636 caffe.cpp:204] Using GPUs 0
I1223 16:55:36.035459 16636 caffe.cpp:209] GPU 0: GeForce GTX 1080
I1223 16:55:36.221006 16636 solver.cpp:45] Initializing solver from parameters:
test_iter: 1000
test_interval: 1000
base_lr: 0.0005
display: 50
max_iter: 25000
lr_policy: "step"
gamma: 0.1
momentum: 0.9
weight_decay: 0.0005
stepsize: 10000
snapshot: 10000
snapshot_prefix: "mscnn_kitti_train_2nd"
solver_mode: GPU
device_id: 0
random_seed: 1706
net: "trainval_2nd.prototxt"
train_state {
level: 0
stage: ""
}
weights: "mscnn_kitti_train_1st_iter_10000.caffemodel"
I1223 16:55:36.221540 16636 solver.cpp:102] Creating training net from net file: trainval_2nd.prototxt
[libprotobuf ERROR google/protobuf/text_format.cc:274] Error parsing text-format caffe.NetParameter: 23:23: Message type "caffe.LayerParameter" has no field named "image_gt_data_param".
F1223 16:55:36.224159 16636 upgrade_proto.cpp:90] Check failed: ReadProtoFromTextFile(param_file, param) Failed to parse NetParameter file: trainval_2nd.prototxt
*** Check failure stack trace: ***
@ 0x7f5a1107f5cd google::LogMessage::Fail()
@ 0x7f5a11081433 google::LogMessage::SendToLog()
@ 0x7f5a1107f15b google::LogMessage::Flush()
@ 0x7f5a11081e1e google::LogMessageFatal::~LogMessageFatal()
@ 0x7f5a1183dbf1 caffe::ReadNetParamsFromTextFileOrDie()
@ 0x7f5a116bb0c5 caffe::Solver<>::InitTrainNet()
@ 0x7f5a116bc425 caffe::Solver<>::Init()
@ 0x7f5a116bc73f caffe::Solver<>::Solver()
@ 0x7f5a118592b1 caffe::Creator_SGDSolver<>()
@ 0x40a778 train()
@ 0x407568 main
@ 0x7f5a0f815830 __libc_start_main
@ 0x407e39 _start
@ (nil) (unknown)
Aborted (core dumped)
this is my first training a model , and I need your help, thanks! @zhaoweicai