Crowd Density Forecasting – CNN + LSTM over YOLO & CSRNet
Extends YOLO+CSRNet crowd counting with a CNN+LSTM pipeline to forecast zone-level crowd density 5–10 minutes ahead and trigger proactive congestion alerts.
Read More →Zulqarnain Haider
I build scalable, impactful solutions in deep learning and NLP, fine-tune transformer models, and design MLOps pipelines for seamless model deployment.
Latest Posts
Open Polymers – Predicting Polymer Properties with ML
ML pipeline using RDKit, PyTorch, and XGBoost to predict Tg, FFV, Density, Tc, and Rg from SMILES strings. Ranked 656/2,400+ (65th on private leaderboard with XGBoost).
Read More →Najm POC – Car Damage Detection using YOLOv12 & ResNet34
A POC that segments car parts with YOLOv12, classifies damage with a fine-tuned ResNet34, and generates a PDF damage report via Streamlit.
Read More →Fine-Tuning a Code LLM on Custom Code with a Single GPU
Fine-tuning large language models (LLMs) for code generation, such as Codex, StarCoder, and Code Llama.
Read More →Unraveling the Power of Knowledge Graphs part-1
Enter the knowledge graph, a transformative tool reshaping the way we navigate and understand information.
Read More →BLEU Metric in NLP Evaluation
Unraveling the Power of BLEU Metric in NLP Evaluation. Evaluating the quality of these generated texts is crucial.
Read More →