jmsquare Goto Github PK

applied-ml

📚 Papers & tech blogs by companies sharing their work on data science & machine learning in production.

augly

A data augmentations library for audio, image, text, and video.

awesome-machine-learning-quality

Coming soon...

awesome-ml-model-governance

This repository provides a curated list of references about Machine Learning Model Governance, Ethics, and Responsible AI.

blog-scraping

This repository provides some R functions to scrap the content of a blog whose URL is organized by date

canonical-correlation-analysis

This project implements canonical correlation analysis between two data matrices. I first create the latent dimensions between the two data matrices. Then I use Kmeans and hierarchical clustering on principal component to group individuals using the latent dimensions and the distance created by the canonical analysis. Last step, I give a profiling of the different groups using descriptive statistics and provide an automated method to export the results.

codecarbon

Track emissions from Compute and recommend ways to reduce their impact on the environment.

documentation

Giskard public documentation

fairness-metrics

This repository provides some group fairness metrics to Machine Learning classifier of German Credit Scoring Dataset. It computes demographic parity, equal opportunity and equalized odd for the sensitive variable gender.

giskard-deployment

Using Giskard's test in Github actions to deploy an AI model.

graph-mining

This project conducts graph mining using R. First it computes the minimum spanning tree and some classical graph mining statistics (density, length, reciprocity, etc). It then builds some centrality measures.

healthcare-analytics

Hospital admission data was analyzed to accurately predict the patient’s Length of Stay at the time of admit so that the hospitals can optimize resources and function better.

itunes

This repository parses music from itunes as a xml file to create a structered SQL database

latent-dirichlet-allocation

This project uses latent dirichlet allocation for topic modeling. It first generates the document term matrix. Then It builds the latent dirichlet allocation algorithm to extract latent topics in the documents. A wordcloud function is also implemented to display the representative words of each topics.

optical-character-recognition

This repository provides 2 functions to read contents and metadata from image pdf files (read.ocr) and from Word document (read.docx). Read.ocr function uses tesseract method to make optical character recognition (OCR) on image pdf file. Read.docx function unzips .docx file to convert to xml file and extract data.

privacy

This work implements some privacy attacks and mitigaton strategy to credit default prediction

sentiment-analysis

spoon-knife

This repo is for demonstration purposes only.