Atlantic: Automated Data Preprocessing Framework for Machine Learning

This script is used to augment image data created using LabelMe-MIT.

A convolutional neural network implemented with tensorflow and trained to recognzie the face of famous persian celebrities.

computational-pathology-pipeline

Real-Time BCI for Rock-Paper-Scissors: Decoding Motor Imagery with Minimal Training

A Python based graphical tool to preprocess tabular datasets

preprocessing for VITON models

Exploring Brain Signal Processing Pipelines for Kaggle Challenges

This repo includes a generalized preprocessing pipeline for text data in NLP tasks.

This repo is provided pipeline for preprocessing fmri & dti data to functional and structural connectivity matrices

Experimental repository for preprocessing evaluation, enhanced from "Action-Based Conversations Dataset: A Corpus for Building More In-Depth Task-Oriented Dialogue Systems"

AutoAD - A framework for the rapid detection of anomalies in (big) datasets

End-to-end data analysis of HR datasets, focusing on insights and trends.

VAPOR - Video Analysis Processing for Object Recognition is a modular framework for analyzing video quality and its impact on medical reconstruction. VAPOR applies controlled blur effects to quantify relationships between image quality metrics and 3D reconstruction accuracy, removes specularity and addresses floating objects in intrabody images.

ImageAndMaskDatasetBuilder is a powerful utility class that streamlines the preprocessing of image-mask datasets for machine learning workflows.

This is a jupyter notebook with all the basic and necessary steps for dwi preprocessing with CSD. This is basically a transcript/notebook version of Dr. Andrew Jahn's YouTube tutorials (https://youtube.com/playlist?list=PLIQIswOrUH68Zi9SVDAdcUExpq2i6A2eD&feature=shared).

This project builds a calibrated Random Forest classifier to predict whether a client will subscribe to a term deposit following a direct marketing campaign. It uses a clean, reproducible pipeline with feature engineering, stratified cross-validation & tuning, and probability calibration.