susansagwa Goto Github PK

ai-ml-task

Artificial Neural Network code that trains the Network in the shared class Notebook found on eLearning by updating its weights using the delta rules of error computation found in the class lecture notes. Output the final trained weight matrices after 100 Epochs

ai-search-strategies

Python Implementation of Search Strategies for Route Finding Problem

calculator

A node.js demo application

coa-assignment-2

Number Systems Usage in Computer Systems

coding-challenge

compiler-construction

Flex, Bison, Lex and Yacc (Description, diagram and code snippets)

computer-graphics-polygons

2D House using OpenGL C++

deep-forward-networks

dresswme

Efashion Website

energy-consumption-prediction-in-smart-buildings-using-ensemble-learning

Building energy consumption prediction plays an irreplaceable role in energy planning and conservation as it is an important part of a building's lifecycle management process, since it allows organizations to make decisions about what investments need to be made in order to reduce energy consumption. Data-driven approaches, such as artificial neural networks, support vector regression, support vector machine, KNN, gradient boosting regression, and extreme learning machine are the most advanced methods for building energy prediction. However, owing to the high nonlinearity between inputs and outputs of building energy consumption prediction models, these approaches require improvement regarding the accuracy of predictions, robustness (i.e., the ability to produce accurate results in situations where there is a high degree of uncertainty or variability), and generalization ability (i.e., the ability to accurately predict outcomes when presented with new data). To counter these shortcomings, an ensemble learning method is proposed, which will use both SVR and SVM. It combines these individual models together to improve the stability and predictive power of the model through learning several simple models and combining their output to produce the final decision.

iap-1-2

Login, Registration,Password Changing and Database Connection

icy

linearregressiontask

Problem Definition: You are expected to generate a fake Nairobi Office Price Simulated dataset of size 100 with One feature 𝒙 (office size assume a suitable distribution) and one target 𝑦 (office price also assume a suitable distribution). Write class logic for Mean Squared Error to be used as your Performance Measure Technique and Gradient Descent as your learning algorithm Set random initial values for slope (m) and y-intercept (c) and train an intelligent linear regression model of your dataset above then plot the line of best fit Use the above learnt line to predict the office price when the size is 100 sq. ft.

mobile-application-development

nisashare-application

A Web Based Book Sharing Application

resume

streamlit-example

Example Streamlit app that you can fork to test out share.streamlit.io

woodland

Woodland is a single player, where each player’s objective is to cut as many timber as he/she can without the branches falling on them. A player’s time will be limited but the more he/she cuts the more the time is extended. The game is supposed to end if time lapses, or the player is hit by a branch. The game served to make the players more active and energetic as it is a stress reliever. The cooperation between the players to chop down as many trees as available on the screen serves to improve the bond between the players, relieve stress and help people going through difficult situations to calm down. It is a fun game that can be played by family or among friends thus it is not specific to a certain age group or individuals. There is only a single character, who can be customized by the player. Scores are earned from the number of chops made by the player.