search

openvinotoolkit / training_extensions

Train, Evaluate, Optimize, Deploy Computer Vision Models via OpenVINO™
https://openvinotoolkit.github.io/training_extensions/
Apache License 2.0
1.14k stars 443 forks source link

tensorflow_toolkit/lpr/ No module named 'spatial_transformer' #325

Closed TNTWEN closed 4 years ago

TNTWEN commented 4 years ago

train error :"openvino_training_extensions/tensorflow_toolkit/lpr/lpr/trainer.py", line 21, in from spatial_transformer import transformer ImportError: No module named 'spatial_transformer'

Missing file spatial_transformer!!!

AlexanderDokuchaev commented 4 years ago

HI @TNTWEN Looks like something go wrong when you run prepare_modules.sh. Please inspect logs. Possible you have no installed 2to3 or protobuf-compiler.

TNTWEN commented 4 years ago

oh !I ignored sudo apt install 2to3 protobuf-compiler and now it works correctly @AlexanderDokuchaev I also want to ask what kind of license plate positioning method should I choose if I use this license plate recognition.I used YOLOv3 in the previous code to identify the car. License plate recognition will be an extension of my code. Can I use YOLOV3 to identify the car and then use opencv to locate the license plate?Or is there any good method i could use in OpenVINO Thanks!

AlexanderDokuchaev commented 4 years ago

You can try to use vehicle-license-plate-detection-barrier-0123 model.

TNTWEN commented 4 years ago

Thanks！i use this vehicle-license-plate-detection-barrier-0123 model and write a python demo.It works great！！