AI/Deep Learning Courses
Registration Form: https://forms.gle/cxunwyhdDea3DEzLA
Getting Started with Deep Learning
Duration: 8 Hours
By participating in this course, you will:
- Learn the fundamental techniques and tools required to train a deep learning model
- Gain experience with common deep learning data types and model architectures
- Enhance datasets through data augmentation to improve model accuracy
- Leverage transfer learning between models to achieve efficient results with less data and computation
- Build confidence to take on your own project with a modern deep learning framework
Click here to go to the NVIDIA website for full details.
Getting Started with Image Segmentation
Duration: 2 Hours
Image (or semantic) segmentation is the task of placing each pixel of an image into a specific class. Learn how to segment MRI images to measure parts of the heart by:
- Comparing image segmentation with other computer vision problems
- Experimenting with TensorFlow tools such as TensorBoard and the TensorFlow Keras Python API
- Learning to implement effective metrics for assessing model performance
Click here to go to the NVIDIA website for full details.
Optimization and Deployment of TensorFlow Models with TensorRT
Duration: 2.5 Hours
In this course, you will learn how to optimize TensorFlow models for more performant inference with the built-in TensorRT integration, called TF-TRT. By the end of the course you will be able to:
- Optimize Tensorflow models using TF-TRT
- Increase inference throughput without meaningful loss in accuracy by using TF-TRT to reduce model precision to FP32, FP16, and INT8
- Observe how tuning TF-TRT parameters affects performance
Click here to go to the NVIDIA website for full details.
Deep Learning at Scale with Horovod
Duration: 2 Hours
Modern deep learning challenges leverage increasingly larger datasets and more complex models. As a result, significant computational power is required to train models effectively and efficiently. In this course, you will learn how to scale deep learning training to multiple GPUs with Horovod, the open-source distributed training framework originally built by Uber. Over the course of 2 hours, you'll:
- Complete a step-by-step refactor of a Fashion-MNIST classification model to use Horovod and run on four NVIDIA V100 GPUs
- Understand Horovod's MPI roots and develop an intuition for parallel programming motifs like multiple workers, race conditions, and synchronization
- Use techniques like learning rate warmup that greatly impact scaled deep learning performance
Click here to go to the NVIDIA website for full details.
AI Workflows for Intelligent Video Analytics with DeepStream
Duration: 2 Hours
The DeepStream 3.0 framework features hardware-accelerated building blocks of Intelligent Video Analytics (IVA) applications. This allows developers to focus on building core deep learning networks. The DeepStream SDK underpins a variety of use cases and offers flexibility on the deployment medium.
You’ll learn how to:
- Deploy DeepStream pipeline for parallel, multi-stream video processing and deliver applications with maximum throughput at scale
- Configure the processing pipeline and create intuitive, graph-based applications.
- Leverage multiple deep network models to process video streams and achieve more intelligent insights
Click here to go to the NVIDIA website for full details.
Modeling Time Series Data with Recurrent Neural Networks in Keras
Duration: 2 Hours
Recurrent neural networks (RNNs) allow models to classify or forecast time-series data, such as natural language, markets, and even patient health care over time. In this course, you'll use data from critical-care health records to build an RNN model that provides real-time probability of survival to aid health care professionals in critical-care treatment decisions.
You'll learn how to:
- Create training and testing datasets from electronic health records in HDF5 (hierarchical data format version five) format
- Prepare datasets for use with RNNs, using normalization, gap-filling, and sequence-padding techniques
- Construct and train a model based on a long short-term memory (LSTM) RNN architectecture, using the Keras API with TensorFlow, then compare the model performance against traditional baseline models
Click here to go to the NVIDIA website for full details.
Medical Image Classification Using the MedNIST Dataset
Duration: 2 Hours
Get a hands-on practical introduction to deep learning for radiology and medical imaging. You'll learn how to:
- Collect, format, and standardize medical image data
- Architect and train a convolutional neural network (CNN) on a dataset
- Learn introductory techniques in data augmentation
- Use the trained model to classify new medical images
Click here to go to the NVIDIA website for full details.
Image Classification with TensorFlow: Radiomics - 1p19q Chromosome Status Classification
Duration: 2 Hours
Thanks to work being performed at the Mayo Clinic, using deep learning techniques to detect Radiomics from MRI imaging has led to more effective treatments and better health outcomes for patients with brain tumors. Learn to detect the 1p19q co-deletion biomarker by:
- Designing and training convolutional neural networks (CNNs)
- Using imaging genomics (radiomics) to create biomarkers that identify the genomics of a disease without the use of an invasive biopsy
- Exploring the radiogenomics work being done at the Mayo Clinic
Click here to go to the NVIDIA website for full details.
Coarse to Fine Contextual Memory for Medical Imaging
Duration: 2 Hours
Coarse-to-Fine Context Memory (CFCM) is a technique developed for image segmentation using very deep architectures and incorporating features from many different scales with convolutional Long Short Term Memory (LSTM). You will:
- Take a deep dive into encoder-decoder architectures for medical image segmentation
- Get to know common building blocks (convolutions, pooling layers, residual nets, etc.)
- Investigate different strategies for skip connections
Click here to go to the NVIDIA website for full details.
Data Augmentation and Segmentation with Generative Networks for Medical Imaging
Duration: 2 Hours
A generative adversarial network (GAN) is a pair of deep neural networks: a generator that creates new examples based on the training data provided and a discriminator that attempts to distinguish between genuine and simulated data. As both networks improve together, the examples created become increasingly realistic. This technology is promising for healthcare, because it can augment smaller datasets for training of traditional networks. You'll learn to:
- Generate synthetic brain MRIs
- Apply GANs for segmentation
- Use GANs for data augmentation to improve accuracy