ERROR: Check failed: fd != -1 (-1 vs. -1) File not found: /home/brp/data/detectnet.prototxt

I am using Digits from host to train Jetson, I get the following error while creating object detection model. I a following this article: https://github.com/dusty-nv/jetson-inference

There was no detectnet.prototxt file. So I created one (pasted the code from: https://github.com/dusty-nv/jetson-inference/blob/master/data/networks/detectnet.prototxt )and dropped it into networks folder.

size-block does not need backward computation.
coverage-block does not need backward computation.
coverage-label_slice-label_4_split does not need backward computation.
obj-label_slice-label_3_split does not need backward computation.
size-label_slice-label_2_split does not need backward computation.
bbox-label_slice-label_1_split does not need backward computation.
foreground-label_slice-label_0_split does not need backward computation.
slice-label does not need backward computation.
val_transform does not need backward computation.
val_label does not need backward computation.
val_data does not need backward computation.
This network produces output loss_bbox
This network produces output loss_coverage
This network produces output mAP
This network produces output precision
This network produces output recall
Network initialization done.
Solver scaffolding done.
Finetuning from /home/brp/data/detectnet.prototxt
Check failed: fd != -1 (-1 vs. -1) File not found: /home/brp/data/detectnet.prototxt

Caffe log output is

I0301 10:06:32.426641  1039 net.cpp:222] slice-label does not need backward computation.
I0301 10:06:32.426645  1039 net.cpp:222] val_transform does not need backward computation.
I0301 10:06:32.426648  1039 net.cpp:222] val_label does not need backward computation.
I0301 10:06:32.426651  1039 net.cpp:222] val_data does not need backward computation.
I0301 10:06:32.426652  1039 net.cpp:264] This network produces output loss_bbox
I0301 10:06:32.426654  1039 net.cpp:264] This network produces output loss_coverage
I0301 10:06:32.426656  1039 net.cpp:264] This network produces output mAP
I0301 10:06:32.426659  1039 net.cpp:264] This network produces output precision
I0301 10:06:32.426664  1039 net.cpp:264] This network produces output recall
I0301 10:06:32.426782  1039 net.cpp:284] Network initialization done.
I0301 10:06:32.427223  1039 solver.cpp:60] Solver scaffolding done.
I0301 10:06:32.429873  1039 caffe.cpp:135] Finetuning from /home/brp/data/networks/detectnet.prototxt
F0301 10:06:32.429888  1039 io.cpp:54] Check failed: fd != -1 (-1 vs. -1) File not found: /home/brp/data/networks/detectnet.prototxt
*** Check failure stack trace: ***
@     0x7fc3f123edaa  (unknown)
@     0x7fc3f123ece4  (unknown)
@     0x7fc3f123e6e6  (unknown)
@     0x7fc3f1241687  (unknown)
@     0x7fc3f17ee62f  caffe::ReadProtoFromBinaryFile()
@     0x7fc3f17f5fb4  caffe::ReadNetParamsFromBinaryFileOrDie()
@     0x7fc3f17c20b7  caffe::Net<>::CopyTrainedLayersFromBinaryProto()
@     0x7fc3f17c2126  caffe::Net<>::CopyTrainedLayersFrom()
@           0x40a604  CopyLayers()
@           0x40ae25  train()
@           0x4086bc  main
@     0x7fc3efd35f45  (unknown)
@           0x408e8d  (unknown)
@              (nil)  (unknown)

Please help me . Thank you in advance

Hi,

The detectnet.prototxt can be installed with the CMakePreBuild.sh script.
Could you uncomment the commands and execute cmake again?

https://github.com/dusty-nv/jetson-inference/blob/master/CMakePreBuild.sh#L56

wget --no-check-certificate 'https://nvidia.box.com/shared/static/xe6wo1o8qiqykfx8umuu0ki9idp0f92p.prototxt' -O detectnet.prototxt
mv detectnet.prototxt ../data/networks

Thanks.

Thank you. Solved after downloading the bvlc goglenet caffe model again.