machine-learning

DistilBERT model with Sequence Classification on 9 newsgroup datasets for the Nepali language with 95.475% accuracy.

About Advanced AI-Generated Image Detection System using EfficientNetB7 with Custom Attention Mechanism. Developed for AICS 2025 (33rd Irish Conference on Artificial Intelligence and Cognitive Science). Detects deepfakes from GANs, Diffusion Models, and other generative AI.

⚡ High-performance multi-model knowledge graph database written in Rust. SQL, Cypher & Gremlin queries · 11 graph algorithms · AI/ML with HNSW vectors · REST API · Zero GC · Blazingly fast.

⚡ Vectro: Lightning-fast embedding quantization. Hit 12M+ vec/s throughput (4.85× faster than FAISS C++) while drastically cutting memory footprint for vector databases and local AI research.

This repo contains evaluation code for MM-IQ benchmark.

Google Earth Engine tool to generate multi-modal and multi-temporal datasets, including spatially and temporally aligned Sentinel-1 SAR data, Sentinel-2 multispectral data, weather and DEM-based data. A supplementary material for Paluba et al. 2024: "Identification of Optimal Sentinel-1 SAR Polarimetric Parameters for Forest Monitoring in Czechia

A new optimizer, Gradient Ascent: Gradient Ascent adjusts the parameters in the direction of the gradient to maximize some objective function.

SkyPortal SDK & CLI - Open Source

Deep learning models(CNN, LSTM, BERT) for image and text classification task with Tensorflow and Keras

Fine-tuning a Large Language Model (LLM) for code vulnerability detection using QLoRA, a method that quantizes the model to 4-bit floats and incorporates adapters for fine-tuning.

Library for handling atomistic graph datasets focusing on transformer-based implementations, with utilities for training various models, experimenting with different pre-training tasks, and a suite of pre-trained models with huggingface integrations

Dataset preparation tool for Mochi-1 video model fine-tuning — clip extraction, captioning & formatting pipeline

This repo shows how to classify crack images and discuss the result with grad-cam