Skeleton-based Self-Supervised Feature Extraction for Improved Dynamic Hand Gesture Recognition

MAE presentation

Modern Analytical Engine

An end-to-end masked contrastive video-and-language pre-training framework

Machine Learning Development - Chicago Weather Time Series

Good Seed were employed Data Science for alcohol law compliance. My role includes using specialized cameras at checkout for alcohol buys, applying advanced computer vision for age verification, and designing a model to confirm age. I built a model with ResNet50 and 'relu', using a single neuron to output.

VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training

R visualisation of output from Viper

The objective of this project is to predict the prevailing wage that is at optimum.

Wine Quality Red Test

Evaluate Video Salient Object Detection Via Python And Cuda !

Investigating Gradient Descent behavior in linear regression

This machine learning project focused on predicting food delivery times. The code emphasizes essential tasks such as data cleaning, feature engineering, categorical feature encoding, data splitting, and standardization to establish a solid foundation for building a robust predictive model.

time series-Jena Climate

Predicting Walmart Sales and Performing Exploratory Data Analysis

In this case study, we have dataset of a company for case study to find actualy proudcity of its workers using regression models. I draw a comparative analysis of different machine learning algorithms on the basis of mean absolute error.

Adversarial Patch defense using SegmentAndComplete (SAC) & Masked AutoEncoder (MAE)

Implementation of a simple linear regression with single feature