sergiomsilva / alpr-unconstrained

License Plate Detection and Recognition in Unconstrained Scenarios

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

what environment did you guys use?

chen3082 opened this issue · comments

I tried to copy Ubuntu 16.04 machine, with Keras 2.2.4, TensorFlow 1.5.0, OpenCV 2.4.9, NumPy 1.14 and Python 2.7 or Python3.5. However, I encountered a lot of problems.

First, installing OpenCV2.4.9 is a pain. I'm not even sure did I do this correctly(I downloaded the zip file). How did you guys do it?
I tried to import cv2 in python2.7. It shows "ImportError: numpy.core.multiarray failed to import”

Second, although I set the default python as 2.7 by alias python=python2 at the top of the ~/.bashrc . Every time I tried to run "run.sh", it kept showing missing parentheses for "print". And every time I tried "python --version" in the terminal, it shows python2.7

Third, I tried to install "yolo34py" as python wrapper by "pip install yolo34py". It also shows error.

I'm wondering how you guys set up the environment. Any help would be appreciated!!!! Thanks!

In ubuntu18.04,I can run this project successfully.this is what I have done
conda create -n alpr-unconstrained python=2.7
conda activate alpr-unconstrained
pip install keras==2.2.4
pip install tensorflow==1.5.0
conda install -c menpo opencv

@ulanbator Thank you sir :)