Car Detection and Counting Project

In this post I'll tell about conflicting issues in vehicle detection and counting project on GitHub and its sync and run on Windows.
Original project link is here = https://github.com/ahmetozlu/vehicle_counting_tensorflow.This project written by Python language on Linux OS.

There are the versions of my PC OS ,Python Language and some Python's Libraries
Python 3.5
Windows 10 Pro
Opencv 3.4.1.15
Numpy 1.14.5
TensorFlow 1.8.0
Scipy 1.1.0

I used to Visual Studio Code as code editor

After the open the original project link click download to project then,at debugging there are the conflicting issues and solutions in ongoing lines step by step:

1.problem = TabError: inconsistent use of tabs and spaces in indentation
Solution = on the Visual Studio Code click "Alt + Shift + F"(converting tabs into spaces)

2. problem = No module name "image_utils"
Solution = I had the to add the utils directory and subfolders PYTHONPATH

3. problem = Cannot import module name "image_saver"
Solution =Let's say I have:

myapp/__init__.py

myapp/myapp.py

myapp/myapp/SomeObject.py

myapp/tests/TestCase.py

              Now if we are trying to get something like this
myapp.py
===============
from myapp import SomeObject
#operations...

TestCase.py
===============
from myapp import SomeObject
#tests

We will see that error
ImportError: No module named 'myapp'

Cause of that routine problem of Python I moved some utilies files path in utils file.

change image_saver.py path then change import lines

4.problem = Some modules same problem(no module name or cannot import module)
Solution = change the .py files path on utils folder

5.problem =Some lines in  knn_classifier.py(While running the program can see video just 3 seconds then disappear)
Solution = it depends on the Python version; read the files not "rb", be "r"  and
not "all_possible_neighbors.iteritems()" ==> "all_possible_neighbors.items"

My project(forked and with commits) project link is here = https://github.com/OnurBerkayDogan/vehicle_counting_tensorflow