Nowadays, with the rapid development of AI, the demand for its use is increasing, leading to a growing abundance of data and a wider variety of tasks. This makes problems in Machine Learning and Deep Learning more time-consuming and memory-intensive to process. In this series, I will guide you on how to optimize Machine Learning and Deep Learning tasks by using GPUs, aiming to enhance performance and efficiency
-
GPU in ML/DL will help you handle Machine Learning and Deep Learning tasks more effectively in all aspects (accuracy, speed, and memory efficiency), from basic to advanced levels, and it will be completely understandable.
-
In this series, you will learn how to optimally and appropriately use GPU when applying them to AI in general, and Machine Learning and Deep Learning in particular. Rest assured, there will be code from scratch as well as guidance on setting up the necessary environments and packages.
- You have to install Cuda Toolkit - here is a guide
- You have to install CuDNN - here is a guide
- You have to install Pytorch (with GPU) - at here
this series focuses on Machine Learning and Deep Learning, you should have a background in:
- Machine Learning: A good understanding of Classification, Regression, and Clustering
- Deep Learning: A good understanding of CNN (Convolutional Neural Network) and Gradient Descent
Please note that for Deep Learning, I will be using PyTorch exclusively (the reasons for choosing PyTorch over TensorFlow will be explained in future posts)
- Chapter01: RAPIDS and What You Should Know
- Chapter02: RAPIDS in handle data
- Chapter03: cuML for Machine Learning
- Chapter04: TPOT AutoML + cuDF