ARgorithmToolkit 0.0.10

A utility toolkit to help develop algorithms suitable for ARgorithm

ARgorithm

Work in Progress

The ARgorithm project provides an interface to render your algorithms and data structures in augmented reality. The ARgorithmToolkit package offers packages and a command line interface needed to make and submit algorithms for the following.

How does it work ?

The Toolkit package is for developers who want to transport their own algorithms into augmented reality. The toolkit provides you with a template library which works just like your usual template library except this one records states . Each state is an event that occurs in your data structure and by keeping track of the states of your variables , data structures etc we then render them in augmented reality.

Getting started

You can install the ARgorithmToolkit using pip

pip install ARgorithmToolkit

or you can clone the repo

git clone https://github.com/ARgorithm/Toolkit.git 
cd Toolkit
make init

This will also setup ARgorithm on your command line

user@pc:~$  ARgorithm -h
usage: ARgorithm [-h] {init,submit,test,delete,account} ...

ARgorithm CLI

optional arguments:
  -h, --help            show this help message and exit

command:
  {init,submit,test,delete,account}
                        try command --help for more details

After that you can get started with your own ARgorithm

ARgorithm init

This will generate your .py file and your .config.json file.

1. The <name>.py file will store your code that you will submit to the server hosting all ARgorithms
2. The <name>.config.json stores important details about your ARgorithm such as its purpose and parameters required

The <name>.py file initially looks like this

import ARgorithmToolkit

def run(**kwargs):
    stateset = ARgorithmToolkit.StateSet()

    #
    #	Your code
	#

    return

You can add whatever code you want to this file using all the tools and classes present in ARgorithmToolkit but be sure to

1. Your file should have one function which takes **kwargs input (refer here to know more about kwargs) that will should perform whatever you desire and should return the stateset. You can check out later in the document on how to use this stateset
2. you can create classes and additional functions in your code. Support for importing external modules is not yet added so its advisable not to add those.

the <name>.config.json file is a JSON file storing all the metadata

{
    "argorithmID" : "<name>",
    "file" : "<name>.py",
    "function" : "<function to be called>",
    "parameters" : {
        "variable-name" : "<data-type>"
    } , 
    "default" : {
        "variable-name" : "<value>"
    },
    "description" : "Tell us about your ARgorithm"
}

Key	Description
argorithmID	name of your ARgorithm , this is generally pre filled from when you run the init command. The name of your code file should be name.py and the config should be name.config.json. [will be fixed later]
file	The file containing your codefile
function	The function that is going to be called
parameters	the parameters that your ARgorithm would need, this helps in anyone using your ARgorithm to understand what is the input format
default	default parameters in case no parameters are passed
description	The description of ARgorithm. Helpful to people using your ARgorithm as well as other developers

You can check out ARgorithm examples in our Github Repo

Once you are done , you can submit to server by running

ARgorithm submit

or

ARgorithm submit --name <name>

you can test your ARgorithm submission by using

ARgorithm test

you can delete your ARgorithm submission by using

ARgorithm delete

If the server need authentication for any action , you will be prompted for it. If you do not have a account on the server then you can register using

ARgorithm account new

you can login using

ARgorithm account login

The access token generated during login stays in your system till it expires so you dont have to login again and again. In case the token expires or is invalid , you will be prompted to login again.

you can use the -o or --overwrite flag to overwrite a pre-existing login

if running server image on local machine , add -l or --local flag in the submit,delete , accountand test commands . To run the server locally , pull the docker image alanjohn/argorithm-server:latest and run it. Check out server repo here

Using ARgorithmToolkit

ARgorithmToolkit adds a few extra steps when it comes to initializing instances whose states you want to record but thats because a lot of data has to be maintained in order for smooth transitions

>>> import ARgorithmToolkit
>>> algo = ARgorithmToolkit.StateSet()
>>> x = ARgorithmToolkit.Variable('x',algo,0,comments='Our first variable')
>>> x.value
0
>>> x.value += 10
>>> x.value
10
>>> print(algo)
{'state_type': 'variable_declare', 'state_def': {'variable_name': 'x', 'value': 0}, 'comments': 'Our first variable'}
{'state_type': 'variable_highlight', 'state_def': {'variable_name': 'x', 'value': 10}, 'comments': ''}

As ARgorithm is tool for creating visual demonstration , you can add comments parameter to most functions. These comments get included in states and get featured as text when that state is rendered in AR.

Make sure you make the objects you want to keep track of as part of the same stateset. Each object is instantiated with a name this is important to identify arrays when rendering them

You can refer the docs and samples in the repo to understand more clearly.

StateSet

The core class to all algorithm and data structures in ARgorithmToolkit You will always need to declare this and use this when using different ARgorithmToolkit features. This is where the states are stored that later get rendered to ARgorithm App. So obviously all your ARgorithms are supposed to return this

You wont have it to interact with it much other than while initialising objects and returning results. It has a add_comment method that allows you to create blank states carrying description in the form of comments that you might want to show to the client while the ARgorithm is being rendered. This will prove handy when creating good demonstrations

>>> import ARgorithmToolkit
>>> algo = ARgorithmToolkit.StateSet()
>>> algo.add_comment("Hello world")
>>> print(algo)
{'state_type': 'comment', 'state_def': None, 'comments': 'Hello world'}