Discover wecet/Human-Activity-Recognition-From-Smartphones Open Source

Stars
5
Rank 2,861,937 (Top 57 %)
Language
Jupyter Notebook
Created over 3 years ago
Updated over 3 years ago

wecet

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Using Android Phone inertia sensors such as gyroscopes and accelerometers to determine and identify human activity. Classified using SVMs, KNN, Decision Trees and Logistic Regression for Supervised Learning

Experimenations-in-Reinforcement-Learning

Experiment 1: Comparison of key bandit algorithms; Experiment 2: Comparison of Q and SARSA Learning on Taxiv3 environment' ; Experiment 3: Comparison of Q, SARSA and CEM Learning on LunarLanderv2 Environment

Jupyter Notebook

Vehicle-Detection-and-Classification

Using YOLOv3, YOLOv4, Faster R-CNN and HoG to detect and classify different types of vehicles from one another

Jupyter Notebook

Comparison-and-Review-of-Pose-Estimation-Implementations

Comparing the performance of the BlazePose and the MoveNet-Thunder frameworks in the tackling of the Pose Estimation problem

Jupyter Notebook

TeaLang-Compiler

A compiler consisting of lexical analysis, syntactic analysis, semantic analysis and interpreting alongside an XML generation for a 'mock' language called TeaLang written in C++14.

C++

Phoneme-Analysis-and-Classification

Given a very large dataset of speech from the British Isles, all were manually inputted into the .csv file attached by generating an average value for each of the 3 formants present in the audio. The scope of the project was to be able to classify the phonemes from one another. Therefore, a KNN was built in order to do just that. Resulting in an average accuracy of 89%

Jupyter Notebook

wecet/Human-Activity-Recognition-From-Smartphones

wecet

Reviews

Repository Details

More Repositories