evanharwin

Vector Field Visualisation Using Time-Based Particle Simulation

Using three.js and Charlie Hoey's GPU Particles ( file: particles.js )

Takes a function that describes a Vector Field over a given domain. Then spawns particles in this domain with a velocity given by the output of the function when given the particle origins.

Also a colour function can be given to output more information. In this case I have used the Curl of the Function.

One should probably note, that the way the particles move over time is in some ways a little misleading as it can suggest a shifting field, rather than a constant vector field.... it's pretty tho.

Also, might have to allow local file loading in browser for this to work.

evanharwin

A Simple Raytracing 3D Graphics Engine

Python - Definitely Not fast But ¯\(ツ)/¯

Written to get back into Mathematics after 4 months Uni Summer Break. Also an awesome thing to impliment, if impractical.

evanharwin

Venmo Data Scraper - I found out about Venmo's public API and decided to dig deeper.

Oh and then got sidetracked by a frustation with Python peewee ( http://docs.peewee-orm.com ) so started to work on storing data with JSON's with great success but a new-found appreciation for the robust-ness of SQL.

However, I was eventually pulled away from this by a huge volume of 429 ( too many requests ) responses from venmo.com/api/v5/public - given the infrequency of my requests, I wonder if this is a deliberate move from Venmo given the bad press they got due to the sources below.

inspired by why-i-blasted-your-drug-deals-on-twitter and public-by-default

evanharwin

A Simple three.js Vector Projection Demo

Made because I needed to learn the formula for Vector Projection and I obviously decided the most ineffient way possible. ... I still remember it, however, so it must have worked ༼ つ ◕_◕ ༽つ

evanharwin

neural.py

Work-In-Progress A machine learning network that can 'sometimes' diagnose Breast Cancer given measurements of potential tumor. Data taken from https://archive.ics.uci.edu/ml/machine-learning-databases/breast-cancer-wisconsin/

evanharwin

binary = "11000000 00000001"

mantissa = binary.split(" ")[0]
exponent = binary.split(" ")[1]

def denary(binary, point_pos):
    total = 0
    index = -(point_pos)
    for bit in binary[-1::-1]:
        total += int(bit) * (2**index)

evanharwin

# The Data Class, currently this shall represent all numbers and be the base of the Var class. 
class Data:
	def __init__(self, value):
		self.value = value

# The Var Class, this shall be a variable with a name and attributed value.
class Var(Data):
	def __init__(self, name):
		self.name = name

evanharwin

#include <iostream>
#include <math.h>

int main() {

	int radius = 26144;
	long int area = 0;
	int x = radius;
	double max_y = radius/sqrt(2);
	double distance;

evanharwin

### for rotation gonna need to change the SDL renderer class. Fun. Needs doing and document.

import time, timeit

bodies_with_velocities = [] # a definitve list of anything and everything that moves 

rigid_body_list = []

# class Quadtree: