This the continuation to the first set of Lagrangian problems/solutions. While it might be a bit foolish to solve such problems, there are a few reasons why they were done: Its good practice to solve some problems that are a little bit different from the norm. It’s good to be forced to think in different ways …
This problem is fairly simple to describe… given a bunch of variables that describe the various aspects of a house along with a sales price, is it possible to come up with a mathematical model that can be used to predict future prices based on different values of the same variables? It is based housing sales data …
First some of the motivation behind this this small project: A while ago I did some optimization work and this project is to ‘awaken some knowledge’ so I can eventually work on some more complicated optimization projects When I did do some optimization work, I had a lot of difficulties with different solvers on the …
First the overall goal should be mentioned… the aim create a model that will classify small images of clothing items. That is, by learning patterns from a set of 60,000 labelled training images, it is possible to predict the category from 10,000 unseen images. The MNIST dataset is good as an introductory dataset, however there are some …
A little while ago I wrote up some of the key details on making a WordPress based website on Google cloud. In this post, a few additional things are discussed which make a WordPress setup a little nicer in terms of data and visualization. Google Analytics… Google Analytics can be included either with a plugin …
This project was more of a large side project than anything else. Its main purpose was to practice making some interactive graphics using the HTML Canvas. Also, it also served to help remember some of the basics with Lagrangian mechanics. Making this project was rather fun. Depending on the workload, more Lagrangian problems will be …
This is a continuation to the previous linear regression document. However this time the focus is on a few different ways to perform non-linear regression using Javascript and various Python frameworks (such as NumPy, SciKitLearn, and TensorFlow). Quadratic Regression Perhaps this is the first logical step beyond linear regression. A parabola of best fit is …
I should have posted this right after I originally created this site, but I wanted to spend some time ironing out any issues first. This way, this document is somewhat more complete. My motivation in creating this site was that I wanted a place to host various projects and other things that I thought were …
This is more of a reference for myself on the multitude of ways that something as computing simple linear regression can be done with Python, SciKitLearn, TensorFlow1, and TensorFlow2. Also, this is a bit of a stepping stone type of project, where the next steps involve non-linear regression, multi-non linear regression in the form of …
Using an already trained network from a previous project, it is possible to make a tool that predicts what the value of number actually is from a primitive sketch of it. The overall tool is a small web app consisting of few different elements, namely: A pretrained convolutional neural network (CNN) based on the MNIST …