deep-learning

The traditional machine learning models give a lot of pain when we do not have sufficient labeled data for the specific task or domain we care about to train a reliable model. Transfer learning allows us to deal with these scenarios by leveraging the already existing labeled data of some related task or domain. We try to store this knowledge gained in solving the source task in the source domain and apply it to our problem of interest. In this work, I have utilized Transfer Learning utilizing BertForSequenceClassification model. Also tried RobertaForSequenceClassification and XLNetForSequenceClassification models for Fine-Tuning the Model.

This repository provides core code for managing large volumes of video footage, enabling content understanding, automatic tagging, and vector database storage. It integrates multimodal models and LLMs for accurate descriptions and semantic search. A web interface allows visualization.

LLM fine-tuning with LoRA + NVFP4/MXFP8 on NVIDIA DGX Spark (Blackwell GB10)

This repository has been created to keep track of my 5 days progress.

Official Implementation of "Fine-Tuning is Fine, if Calibrated.", NeurIPS 2024

A Deep Learning project that uses Diffusion transformers (DiT) to generate Grand Theft Auto V driving footage

Near-optimal vector quantization from Google's ICLR 2026 paper — 95% recall, 5x compression, zero preprocessing, pure Python FAISS replacement

Why is LLM inference slow — and how do you make it fast? A hands-on, first-principles course: roofline → KV cache → quantization → parallelism → vLLM/SGLang, with GPU labs on open models.

Implementation of build an LLM from scratch by Sebastian Raschka.

⚡ A seamless integration of HuggingFace Transformers & Diffusers with RBLN SDK for efficient inference on RBLN NPUs.

The elegant integration of huggingface/nlp and fastai2 and handy transforms using pure huggingface/nlp

indoBERT Base-Uncased fine-tuned on Translated Squad v2.0

My own playground for PLP (Programming Language Processing) using DeepLearning techniques

Code for reproducing the experiments on large-scale pre-training and transfer learning for the paper "Effect of large-scale pre-training on full and few-shot transfer learning for natural and medical images" (https://arxiv.org/abs/2106.00116)

Implementation of TSDS: Data Selection for Task-Specific Model Finetuning. An optimal-transport framework for selecting domain-specific and task-specific training data to improve LLM finetuning and instruction tuning.

KAREN: Unifying Hatespeech Detection and Benchmarking

End-To-End Machine Learning Project for Automatic Grades for Student Essays

We introduce OpenStory++, a large-scale open-domain dataset focusing on enabling MLLMs to perform storytelling generation tasks.

Grounding Language Models for Compositional and Spatial Reasoning

Multi-Modal Hate Speech Detection using Deep Learning.