If you already know what Git, Python, Tensorflow, and machine learning are, you can skip to the Onboarding section.
The Inertial Sensing Lab project hosts all of its code base on GitHub, a hosting service that is geared towards software development.
The main tool that is used to interact with our code base on GitHub is called Git, an application developed by Linus Torvalds as a way to manage source codes.
Currently our project is Invite Only. To access our code base, prospective users need to register an account on GitHub, and request access from the project director: Chris Larnder.
Python is an interpretive programming language that we use for developing functionalities that aid us in our researches. Project is written in Python v3.6.x up to v3.7.x
Tkinter is a framework for building graphical user interfaces. In our case, it is mostly used for file input and output dialogs platform-agnostically.
Tensorflow is a widely-used machine learning framework originally developed at Google by Andrew Ng. It could be used in conjunction with many programming languages, specifically Python, to train predictive models. We are using Tensorflow major version v1.x.x
Matplotlib is a dynamic plotting graphing library for Python. We use it, of course, for visualization and aid the analysis of our ML model results.
Numpy is a Python library that is used for implementations of linear algebra concepts such as support for vectors, matrices, and N-dimensional arrays. In addition, the library supports high-level math operations.
Terminal
by opening up Spotlight search with ⌘Command + Space and search for Terminal
. Then, hit Entergit
in the Terminal
and hit Enter.Terminal
and the git
command can then be used.v3.7.x
. Later versions might work as well, but it is safest to go with v3.7.x
to avoid unnecessary deprecations and bugs.v3.7.x
.exe
installer file on the next pagex86 for 32-bit Windows version and x64 for 64-bit Windows version
Most likely you are running a 64-bit version of Windows if your computer was bought no earlier than 10 years ago.
Administrative privilege would be needed for a complete installation of Python
During the installation, look out for the option to add “Python to PATH environment variables” and the option to “remove the path length limit” and check them all.
After the installing, you can open up Command Prompt and run the follow command to check if it’s installed correctly.
Input: ($
symbol indicates shell by standard syntax, not needed when typing in commands)
$ python
Example output:
Python 3.8.5 (default, Jul 28 2020, 12:59:40)
[GCC 9.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>>
v3.x.x
should come with xcode command line tools on macOSTerminal
and run this command to check$ python3
Example output:
Python 3.8.5 (default, Jul 28 2020, 12:59:40)
[GCC 9.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>>
v3.6.x
, or v3.7.x
, or v3.8.x
To run our developed tests with PyCharm IDE, there are some initial setups needed to be done first for PyCharm to work correctly with the project code base.
File
→ Open...
this will open up a file dialog<No Interpreter>
to selectInterpreter Settings
+
sign to add the required Python librariesTensorflow
and click on Tensorflow
Specify version
and select Tensorflow 1.15.0 from the dropdownInstall Package
button afterwardsmatplotlib
, numpy
.Interpreter Settings
Project Structure
optionsrc
folder and mark it as Sources$ git clone https://github.com/larnder/2019_06_AccelerationCamp.git islab
C:\Users\your_user_folder\
, so the cloned repository will be at C:\Users\your_user_folder\islab
./Users/your_user_folder/islab
cd
(change directory) command to point to the repo’s file path on your computerExample of cd
usage:
$ cd %USERPROFILE%\islab
$ cd ~/islab
or
$ cd $HOME/islab
pip
module to install the dependencies:$ python -m pip install -r requirements.txt
or
$ python3 -m pip install -r requirements.txt
.md
files with its Markdown plug-in.The project code base has 2 important folders: data
, src
data
folder:src
folder:modules
, tests
modules
folder has lower-level tools to load data, draw graphs, do mathematical transformations, and etc. These are the functions with which tests are built.test_units
folder has debugging and error-catching tools for developmenttests
folder is what you will be running as examplestests
folder to see the results.[email protected]
with relative information for diagnosis and resolution.This section covers common bugs in the process of running the tools developed in this project. Please refer to this first before resolving to contacting the project maintainers.
Before starting the troubleshooting process, please gather some information pertaining to your computer system by using these following commands:
→ This will open up DirectX dialog to view your PC’s hardware specifications which include CPU, GPU, display monitor manifest, RAM, etc. that would help to narrow down the cause of the errors.