CaffeParser: Could not parse deploy file Segmentation fault (core dumped)

So i have been trying to convert caffe models to .engine files in cpp
I keep getting the following error, kindly someone help. Where might be the problem?

[libprotobuf ERROR google/protobuf/text_format.cc:298] Error parsing text-format ditcaffe.NetParameter: 7:1: Expected identifier, got: <
ERROR: CaffeParser: Could not parse deploy file
Segmentation fault (core dumped)

Hi,

Based on the above error text, it seems to be some issue with prototxt file. May be some typo or unsupported element.
Could you please share your script and model file so we can better help?

Meanwhile, could you please try run your caffe models to verify the model files?

Thanks

UNKNOWN: Plugin Creator registration succeeded - GridAnchor_TRT
UNKNOWN: Plugin Creator registration succeeded - NMS_TRT
UNKNOWN: Plugin Creator registration succeeded - Reorg_TRT
UNKNOWN: Plugin Creator registration succeeded - Region_TRT
UNKNOWN: Plugin Creator registration succeeded - Clip_TRT
UNKNOWN: Plugin Creator registration succeeded - LReLU_TRT
UNKNOWN: Plugin Creator registration succeeded - PriorBox_TRT
UNKNOWN: Plugin Creator registration succeeded - Normalize_TRT
UNKNOWN: Plugin Creator registration succeeded - RPROI_TRT
UNKNOWN: Plugin Creator registration succeeded - BatchedNMS_TRT
size48885336
size48885336INFO: Glob Size is 21688832 bytes.
INFO: Added linear block of size 759296
INFO: Added linear block of size 379904
INFO: Added linear block of size 184832
INFO: Added linear block of size 111104
INFO: Found Creator L2Norm_Helper_TRT
INFO: Found Creator L2Norm_Helper_TRT
INFO: Deserialize required 10449577 microseconds.

Start generating mtCNN TenosrRT runtime models
changedH = 96, changedW = 128
rawName = …/mtCNNModels/det1_relu1.engine
[libprotobuf ERROR google/protobuf/text_format.cc:298] Error parsing text-format ditcaffe.NetParameter: 7:1: Expected identifier, got: <
ERROR: CaffeParser: Could not parse deploy file
Segmentation fault (core dumped)

Hi,

Can you share your code/script and model file so I can try to reproduce this or further debug?

Thanks

i was implementing this GitHub - nwesem/mtcnn_facenet_cpp_tensorRT: Face Recognition on NVIDIA Jetson (Nano) using TensorRT on the nano

Hi,

It seems, issue is due to downloaded prototxt files using wget command.
eg: wget MTCNN_FaceDetection_TensorRT/det1_relu.prototxt at master · PKUZHOU/MTCNN_FaceDetection_TensorRT · GitHub

The downloaded files contains the html tags, due to which you are getting following error:
“Error parsing text-format ditcaffe.NetParameter: 7:1: Expected identifier, got: <”

Please download/copy the valid prototxt file.

Thanks

^ The raw version of that file for downloading with wget would be: https://raw.githubusercontent.com/PKUZHOU/MTCNN_FaceDetection_TensorRT/master/det1_relu.prototxt

You have to click on the “Raw” button on the top right when viewing a file in GitHub to get a downloadable link rather than the HTML link.