Build, train, and backpropagate deep neural networks with GPU acceleration, transfer learning, and regularization.
Real-time object detection, segmentation, and facial recognition using OpenCV and state-of-the-art YOLO architectures.
Quantize and optimize deep models using TensorRT and ONNX for real-time deployment on edge devices like Jetson and Raspberry Pi.
This comprehensive workshop equips engineering undergraduates and postgraduate researchers with real-world AI and deep learning competencies. Moving beyond theory, participants train and optimize state-of-the-art computer vision models, build transformer-based NLP workflows, and deploy models onto edge computing hardware.
Vectorized computing with NumPy, data manipulation with Pandas, statistical distributions, gradient descent, and loss optimization.
Regression models, classification algorithms, Random Forests, XGBoost, clustering techniques, and model evaluation metrics.
Multi-layer perceptrons, backpropagation, activation functions, regularization, and training deep neural networks with GPU acceleration.
Image processing fundamentals, Convolutional Neural Networks (CNNs), transfer learning, and real-time object detection with YOLOv8.
Text embeddings, tokenization, transformer architectures, HuggingFace pipeline integration, and fine-tuning LLMs.
Quantization, ONNX / TensorRT conversion, edge inferencing on Raspberry Pi / Jetson, and end-to-end project presentation.
Book an on-campus demonstration, request turnkey lab setup proposals, or customize syllabi for your department.