pytorch

source code and pre-trained/fine-tuned checkpoint for NAACL 2021 paper LightningDOT

Real-time football video analytics: player/ball/referee tracking and pitch keypoints. YOLO, HRNet, OSNet, ByteTrack, FastAPI. Game state recognition with lightweight CLIP/VLM validation. Python, OpenCV, CUDA/MPS.

Unofficial Implementation of the Google Paper - https://sketch-guided-diffusion.github.io/

Reimplementation of a BERT based model (Shi et al, 2019), currently the state-of-the-art for English SRL. This model implements also predicate disambiguation.

Free AI Image Generator API — Nano Banana Pro (Gemini 3.0 Pro). Create ultra-realistic images, edit photos, and generate stunning AI visuals in seconds — 100% free, no watermark. The ultimate AI Image solution for creators and developers.

A deep learning library based on Pytorch focussed on low resource language research and robustness

Code for Video Deepfake Detector from "MINTIME: Multi-Identity Size-Invariant Video Deepfake Detection", paper available on IEEE Transactions on Information Forensics and Security.

Code and data form the paper BERT Got a Date: Introducing Transformers to Temporal Tagging

The htop for LLM inference see exactly where every GB of VRAM goes and get measured quantization savings.

Exploring Japanese SimCSE

Library - Vanilla, ViT, DeiT, BERT, GPT

A PyTorch Lightning extension that accelerates and enhances foundation model experimentation with flexible fine-tuning schedules.

Pytorch implementation of image captioning using transformer-based model.

Applying BERT for named entity recognition on resumes.

A brand logo detection system by DETR

we propose to generate a series of geometric shapes with target colors to disentangle (or peel off ) the target colors from the shapes. By jointly learning on multiple color-shape images, we found that the method can successfully disentangle the color and shape concepts.

The official repository for our paper "The Devil is in the Detail: Simple Tricks Improve Systematic Generalization of Transformers". We significantly improve the systematic generalization of transformer models on a variety of datasets using simple tricks and careful considerations.

Build Recommender System with PyTorch + Redis + Elasticsearch + Feast + Triton + Flask. Vector Recall, DeepFM Ranking and Web Application.

Modality-Transferable-MER, multimodal emotion recognition model with zero-shot and few-shot abilities.

Deep Learning Framework for Image Classification & Regression in Pytorch for Fast Experiments