t-SNE exploration method on MNIST dataset

sciBASIC# is a kind of dialect language which is derive from the native VB.NET language, and written for the data scientist.

Comprehensive Machine Learning Techniques: Metrics, Classifiers, and Evaluation

🚀 Create your own simulation of t-SNE on histopathology images!

Exploration and optimization of a ML pipeline, delving into various techniques for enhancing different stages of ML workflows, including data preprocessing, feature engineering, model selection, and hyperparameter tuning.

NLP with NLTK for Sentiment analysis amazon Products Reviews

ASAP : Automated Single-cell Analysis Pipeline

This repository features a project on handwritten digit recognition using PCA for dimensionality reduction and K-Means Clustering for grouping the MNIST dataset. t-SNE is employed to visualize the clustering results in 2D.

nostr t-SNE graph

Conducted a comprehensive clustering analysis to categorize beers based on features such as Astringency, Alcohol content, Bitterness, Sourness, and more. Utilized k-medoids and hierarchical agglomerative clustering algorithms to achieve this classification. Tech: Python (numpy, pandas, seaborn, matplotlib, sklearn, scipy)

A JavaScript Library for Dimensionality Reduction

Developed an anomaly detection using Autoencoder Neural Networks to identify outliers in datasets. Preprocessed data with feature scaling, designed a deep autoencoder model, and trained it to minimize reconstruction error using MSE. Classified anomalies based on reconstruction error and visualized latent features with t-SNE, achieving high accuracy

SNE Simulation Dataset Functions

R package for dimensionality reduction of small datasets

Python package for plug and play dimensionality reduction techniques and data visualization in 2D or 3D.

This repository explores word vectors in NLP, including tokenization, vocabulary building, and generating vectors with PPMI and GloVe, using t-SNE to visualize semantic relationships.

Stochastic Neighbor Embedding Experiments in R

Implements Decision Trees, Neural Networks, and Random Forests to optimize the Readahead feature in Linux. Includes data preprocessing, feature selection, and model training for dynamic workload classification and system performance improvement.

Implementation of t-SNE with c++ (including the random walk version), visualization of the process of t-SNE on MNIST with python.