Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Course Outline
MATLAB Deep Learning Environment & GPU Validation
- Overview of the Deep Learning Toolbox architecture and operational workflow
- Procedures for verifying GPU availability, ensuring CUDA/cuDNN compatibility, and configuring drivers for government systems
- Configuration of parallel workers, memory management protocols, and foundational use of
gpuArray - Lab 1: Validating the computing environment and executing a initial GPU-accelerated deep learning script
Core Deep Learning Constructs in MATLAB
- Neural network layer components: convolutional, pooling, batch normalization, dropout, residual, and dense layers
- Fundamentals of
dlarray,dlnetwork, and the implementation of custom training loops - Application of loss functions, optimizers (Adam, SGD, RMSProp), and learning rate scheduling strategies for government applications
- Visualization techniques for network architectures, weight distributions, and gradient flow to facilitate debugging
- Lab 2: Constructing a custom
dlnetworkfrom scratch and analyzing layer interactions
Designing CNNs for Image Recognition
- CNN design patterns: feature extraction, spatial hierarchies, and receptive field analysis
- Implementation of transfer learning using pre-trained networks such as ResNet, EfficientNet, and MobileNet
- Data augmentation workflows utilizing
imageDatastore,augmentedImageDatastore, and custom transforms - Lab 3: Training a CNN from inception on a custom image classification dataset with integrated augmentation
Automated Data Labeling & Reproducible Pipelines
- Leveraging MATLAB’s active learning and semi-supervised labeling tools for efficient data processing
- Import and export procedures for annotations (COCO, Pascal VOC, YOLO, CSV) to support interoperability
- Development of version-controlled, parameterized data preparation scripts to ensure auditability
- Lab 4: Automating the labeling workflow and integrating it into a standardized training script
Scalable Training: Multi-GPU, Cloud & Clusters
- Multi-GPU training strategies: batch size optimization, gradient accumulation, and data parallelism for government workloads
- Distributed training execution using MATLAB Parallel Server and on-premises computing clusters
- Cloud training workflows (AWS, Azure, GCP) via MATLAB cloud compute profiles
- Training monitoring, checkpointing protocols, and hyperparameter optimization techniques
- Lab 5: Scaling a model to a multi-GPU or cloud infrastructure and profiling training throughput
Cross-Framework Interoperability & Model Exchange
- Importing pre-trained Caffe and TensorFlow/Keras models into the MATLAB environment
- Validating accuracy parity and adapting architectures for seamless integration into MATLAB workflows
- Exporting models to ONNX, TensorFlow, or Core ML formats to facilitate cross-platform deployment for government partners
- Lab 6: Importing a TF-Keras model, performing fine-tuning in MATLAB, and exporting the result to ONNX
Capstone Project & Production Readiness
- End-to-end pipeline management: data ingestion, training, validation, optimization, and deployment
- Model compression techniques: pruning, quantization, and code generation using GPU Coder
- Reproducibility best practices: logging standards, seed management, and sharing MATLAB deep learning apps within government teams
- Capstone: Designing, training, optimizing, and exporting a complete image recognition system tailored to specific domain requirements
To request a customized course outline for this training, please contact us.
Requirements
- Demonstrated competence in MATLAB, including syntax mastery, programming methodologies, and familiarity with relevant toolboxes
- No previous background in data science or deep learning is necessary
- Availability of a local GPU-enabled workstation (CUDA-compatible) or an approved cloud cluster for participation in live laboratory exercises
Audience
- Developers and Software Engineers
- Research Engineers and Domain Experts
- Teams transitioning from traditional signal and image processing to AI-driven workflows for government applications
14 Hours
Testimonials (2)
The training was organized and well-planned out, and I come out of it with systematized knowledge and a good look at topics we looked at
Magdalena - Samsung Electronics Polska Sp. z o.o.
Course - Deep Learning with TensorFlow 2
I really liked the end where we took the time to play around with CHAT GPT. The room was not set up the best for this- instead of one large table a couple of small ones so we could get into small groups and brainstorm would have helped
